Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Does relative remoteness affect chronic disease outcomes? Geographic variation in chronic disease mortality in Australia, 2002–2006

Does relative remoteness affect chronic disease outcomes? Geographic variation in chronic disease... R ecent studies have reported differences in mortality outcomes between remote and urban populations, with a correlation between increasing remoteness and excess deaths observed in both Australia and overseas. Some researchers have described the scale of discrepancies while others have examined the cause of the differences. Studies in Australia have also highlighted the substantial mortality burden attributable to socioeconomic inequality and proposed that this may, in part, be the result of inequality in the individual benefits of healthcare. Poorer health care access, health risk factors and environmental health hazards have been associated with increasing remoteness. An additional factor that may also explain poorer mortality outcomes for remote areas is the higher concentration of Indigenous people who, as a population, have poorer health outcomes including mortality. Among the Indigenous population, the association of mortality and remoteness has varied, with two studies reporting lower mortality in very remote areas, while in Queensland, Indigenous death rates have recently been reported as higher in remote areas. Past studies investigating geographic remoteness and mortality have examined either Australia as a whole or selected states. In Australia, chronic diseases contribute significantly to premature mortality, but we know of no previous studies that have examined the six leading causes of chronic disease mortality across all states and territories by relative remoteness. Further, in this study we examined national death registration data and dealt with data discrepancies associated with geographic boundary changes that occurred during the study period. Historically, only underlying cause of death was available in death registration data although it is well documented that underlying cause of death alone underestimates the true burden of chronic disease mortality. More recent studies have used multiple causes of death, to more accurately estimate the impact of any one particular disease on mortality. This study examined the variation in multiple cause mortality rates from chronic diseases by relative remoteness for Australia as a whole, and within each state and territory. The study was approved by the Human Research Ethics Committee of the Northern Territory Department of Health and Menzies School of Health Research (HREC, 2011–1551). Methods Death data This study used Australian Bureau of Statistics (ABS) death registration data from 2002 to 2007 to identify deaths that occurred between 2002 and 2006 across all ages. The additional registration year (2007) allowed the inclusion of late registrations. Death registration data contains a deceased person's age; sex; usual residence coded to five‐digit Statistical Local Area (SLA); and disease diagnosis codes for underlying cause of death and associated causes of death. The coding, completed by ABS, used the International Classification of Diseases, 10 th Revision (ICD‐10). Six common chronic diseases were included: diabetes (E10‐E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19). Cases were selected using multiple causes of death (both underlying and associated causes) for the six chronic diseases. In ABS death registration data, usual residence addresses recorded in death certificates were assigned a five‐digit SLA code. These codes were based on the Australian Geographic Classification System (ASGC) SLA codes from the previous year. Historically, SLA geographic boundaries were modified annually to incorporate new local government council areas, expansion of suburban areas or realignment of town planning zones. Based on our analysis, it was estimated an average 6% of total SLAs in Australia were modified between 2002 and 2006. In this study, we mapped the 2002 to 2006 SLA level death data to 2006 SLAs using ABS population weighted correspondence files. Of the total 663,121 death records between 2002 and 2006, the number of deaths excluded from analysis was 2,411 (0.36%), which included deaths that occurred in the Off‐Shore Areas & Migratory SLAs and records with invalid or missing SLAs. Measure of Remoteness Area In 2001, ABS listed the Remoteness Area (RA) classification within the ASGC. The RA classification was based on the Accessibility/Remoteness Index of Australia (ARIA), initially developed by the Commonwealth Department of Health and Ageing (DoHA) and the National Centre for Social Applications of GIS (GISCA). The RA classification was derived using the road distance to the nearest urban centres in each of five categories based on population size. The RA classification defines Australian areas into the five categories: Major Cities (MC), Inner Regional (IR), Outer Regional (OR), Remote (R) and Very Remote (VR) areas. Population data The ABS estimated resident population (ERP) for the years from 2002 to 2006, rebased to 2006 SLAs, was obtained from ABS. The population data contains five‐year age group, sex, SLA code and states and territories identifier. The population data used in this study were consistent with death data after the previous adjustment, by ABS, from 2002 to 2005 SLA level ERPs to 2006 SLAs, using population weighted correspondence files. The SLA level death and population data were aggregated to RA categories by states and territories and Australia, using ABS population weighted correspondence file of 2006 RA to 2006 SLA. The study data, by RA categories, were further aggregated to three analysis groups: all persons; persons aged 35–64; and persons aged 65 years and above. Statistical analysis Age‐standardised mortality rates for the combined six chronic diseases for all persons, persons in the age groups 35–64 and 65 years and above, and rates for each individual chronic disease for all persons were calculated using direct standardisation to the 2001 Australian standard population. Chronic disease mortality rates for each RA category of states and territories and Australia were examined for the two age groups and all persons. A standardised rate ratio (SRR) for chronic disease deaths was used to measure the differences in mortality rate between capital cities of states and territories and very remote areas. All statistical analyses were conducted using STATA software, version 11. Results Overall chronic disease mortality During the period 2002–2006, there were 240,975 deaths with one of the six selected chronic diseases listed as the underlying cause and an additional 114,261 deaths with at least one of the chronic diseases listed as an associated cause of death. Of the total multiple causes of chronic disease deaths (355,236), 10.3% of deaths were in the age group 35–64 years and 89.3% deaths were in the age group 65 years and above, with the small balance (0.4%) among those aged less than 35 years. In the age group 35–64 years, the proportion of total deaths in each of the remoteness categories increased with increase in remoteness in both males and females. As a result, there was a converse distribution in the older age group with the proportion of total deaths by remoteness areas decreasing with increase in remoteness (Table ). Overall, at least one of the six chronic diseases was recorded as a cause of death for more than half (54%) of total deaths. Ischaemic heart disease was the leading cause of chronic disease death and was recorded as either the underlying or associated cause of death for half of total deaths (51%), followed by: stroke (28%); hypertension (18%); chronic obstructive pulmonary diseases (COPD, 17%); diabetes (16%); and chronic renal disease (13%). Number and proportion of chronic disease deaths for five remoteness categories a by age group and gender, 2002–2006, Australia. Male Female Person no % no % no % <35 years Major Cities Inner Region Outer Region Remote Very Remote Australia 581 190 102 41 66 978 0.5 0.5 0.5 1.5 4.8 0.5 367 127 73 26 19 604 0.3 0.3 0.4 1.4 2.0 0.3 948 317 175 67 85 1,582 0.4 0.4 0.5 1.5 3.7 0.4 35–64 years Major Cities Inner Region Outer Region Remote Very Remote Australia 15,303 5,629 3,291 704 573 25,500 13.4 13.5 16.5 26.5 41.4 14.2 6,547 2,507 1,362 328 361 11,105 5.6 6.4 8.2 17.4 38.2 6.3 21,850 8,136 4,653 1,032 934 36,605 9.5 10.1 12.8 22.7 40.1 10.3 65 years & above Major Cities Inner Region Outer Region Remote Very Remote Australia 98,410 35,991 16,541 1,915 744 153,602 86.1 86.1 83.0 72.0 53.8 85.3 109,960 36,285 15,103 1,526 564 163,447 94.1 93.2 91.3 81.2 59.7 93.3 208,370 72,276 31,644 3,441 1,308 317,049 90.1 89.5 86.8 75.8 56.2 89.3 All persons Major Cities Inner Region Outer Region Remote Very Remote Australia 114,294 41,810 19,934 2,660 1,383 180,080 100.0 100.0 100.0 100.0 100.0 100.0 116,874 38,919 16,538 1,880 944 175,156 100.0 100.0 100.0 100.0 100.0 100.0 231,168 80,729 36,472 4,540 2,327 355,236 100.0 100.0 100.0 100.0 100.0 100.0 a The Remoteness Area classification is based 2006 census year . Selected chronic diseases are: diabetes (E10–E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19) . Chronic disease mortality rates by Remoteness Area (RA) at national level Table shows age‐standardised chronic disease mortality rates for age groups: 35–64 years; 65 years and above; and all persons. The rates were lowest among those living in MC and consistently increased toward VR (Table ). This pattern was seen in both males and females and both age groups. A standardised rate ratio was used to compare mortality rates in different remoteness categories with the rate in the MC category. The SRR to MC in the age group 35–64 years is much higher than age group 65 and above. In the age group 35–64, the mortality rate in VR is 4.2 times higher than MC, female rates much higher (6.2 times) than males rates (3.2 times). In the age group 65 and above, the SRR is moderately increased with remoteness; in VR the mortality rate among males is 1.2 times higher than MC and 1.3 times higher among females. Age standardised chronic disease mortality rates (and 95%CI) for five remoteness categories a by age group and gender, 2002–2006, Australia. Male Female Person Rate 95% CI Rate 95% CI Rate 95% CI Deaths per 100,000 population b 35–64 years Major Cities Inner Region Outer Region Remote Very Remote Australia 113.7 129.6 151.8 210.0 368.3 124.7 (111.9–115.5) (126.2–133.0) (146.6–157.0) (194.6–225.6) (338.2–398.5) (123.1–126.2) 47.9 57.9 67.9 117.5 294.5 54.4 (46.7–49.1) (55.6–60.2) (64.3–71.5) (104.8–130.2) (264.1–324.9) (53.5–55.5) 80.5 93.9 111.5 168.1 335.5 89.6 (79.4–81.6) (91.8–95.9) (108.3–114.7) (157.8–178.3) (314.0–357.1) (88.7–90.5) 65 years & above Major Cities Inner Region Outer Region Remote Very Remote Australia 2,721.0 2,892.0 3,011.0 3,090.8 3,258.6 2,796.8 (2,704.0–2738.0) (2,862.2–2922.0) (2,965.1–3056.9) (2,952.4–3229.2) (3,025.0–3492.7) (2,782.8–2810.8) 2,078.0 2,186.2 2,216.2 2,271.0 2,670.0 2,118.2 (2,065.7–2090.3) (2,163.7–2208.7) (2,180.8–2251.5) (2,157.1–2385.0) (2,449.5–2890.4) (2,107.9–2128.4) 2,353.4 2,502.4 2,585.4 2,672.2 2,984.4 2,415 (2,343.3–2363.5) (2,484.1–2520.6) (2,556.9–2613.8) (2,583.0–2761.5) (2,822.6–3146.1) (2,406.6–2423.4) All persons Major Cities Inner Region Outer Region Remote Very Remote Australia 386.7 414.9 438.4 473.4 563.7 400.8 (384.5–388.9) (410.9–418.9) (432.3–444.5) (455.4–491.4) (534.0–593.4) (398.9–402.7) 280.2 297.9 305.9 333.5 452.6 287.9 (278.6–281.8) (295.0–300.9) (301.2–310.6) (318.4–348.5) (423.7–481.5) (286.6–289.2) 327.5 351.8 369.3 404.0 512.4 339.0 (326.2–328.9) (349.4–354.3) (365.5–373.0) (392.3–415.8) (491.6–533.2) (337.9–340.1) a The Remoteness Area classification is based 2006 census year . b Estimated Resident population for the year 2002–2006, rebased to 2006 Remoteness Areas . Selected chronic diseases are: diabetes (E10–E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19) . A similar trend was also observed for individual chronic diseases for all persons, with MC having the lowest mortality rate increasing towards VR (Table ). There was a minor reversal in the point estimates for stroke between IR and OR remoteness categories. Age standardised mortality rates (and 95% CI) for six chronic diseases by remoteness category, a 2002–2006, Australia. Chronic diseases Major Cities (MC) Inner Region (IR) Outer Regional (OR) Remote (R) Very Remote (VR) Australia (AUS) Deaths per 100,000 population b Diabetes 51.6 (51.1–52.1) 55.4 (54.4–56.3) 64.8 (63.3–66.4) 84.1 (78.9–89.4) 139.7 (129.1–150.3) 54.5 (54.1–54.9) Ischaemic heart disease 169.6 (168.6–170.6) 180.1 (178.4–181.8) 183.8 (181.1–186.4) 190.7 (182.6–198.7) 222.3 (208.8–235.7) 174 (173.2–174.8) Stroke 93.3 (92.5–94.0) 95.2 (94.0–96.5) 93.9 (92.0–95.8) 96.5 (90.7–102.4) 106.1 (96.2–116.0) 94 (93.4–94.6) Hypertension 59.7 (59.1–60.3) 60.8 (59.8–61.8) 67.0 (65.4–68.7) 78.2 (73.0–83.4) 96.7 (87.5–105.8) 61.1 (60.6–61.6) Chronic obstructive pulmonary disease 53.6 (53.0–54.1) 62.8 (61.8–63.8) 71.0 (69.4–72.7) 81.3 (76.0–86.5) 113.5 (103.5–123.6) 57.9 (57.4–58.4) Renal 44.3 (43.8–44.8) 45.0 (44.2–45.9) 48.0 (46.7–49.4) 60.3 (55.8–64.9) 92.5 (83.7–101.3) 46.6 (46.2–47.0) a The Remoteness Area classification is based 2006 census year . b Estimated Resident population for the year 2002–2006, rebased to 2006 Remoteness Areas . Chronic disease mortality rates by Remoteness Area (RA) for states and territories Age‐standardised mortality rates for chronic diseases also varied across RAs within state and territories; again there was a general increase in mortality rates with increasing remoteness, with the exception of Victoria (Table ). In Victoria, the point estimate for the R mortality rate was lower than either the IR or OR categories, though there was no statistical difference between the three categories. The greatest increase in mortality rate with increased remoteness was in the NT, where the VR mortality rate was 1.9 times higher than the Darwin area – classified as OR in remoteness categories. Age standardised chronic disease mortality rates (and 95% CI) for States and Territories by remoteness category, a 2002–2006, Australia. Major Cities (MC) Inner Regional (IR) Outer Regional (OR) Remote (R) Very Remote (VR) Deaths per 100,000 population b New South Wales & Australian Capital Territory 336.3 (334.1–338.6) 351.9 (347.9–356) 366.3 (359.2–373.5) 448.9 (415.9–481.9) 490.3 (390–590.5) Victoria c 323.5 (321–326.1) 360.6 (355.7–365.4) 369.6 (360.3–378.8) 349.3 (286.8–411.7) Queensland 328.4 (325–331.7) 352.1 (346.6–357.6) 367.7 (360–375.3) 409.2 (385.6–432.8) 479.5 (446.6–512.5) South Australia 323.9 (319.7–328.1) 311.4 (300.8–322) 379.8 (368.3–391.3) 360.6 (337.1–384.1) 451.5 (395–508.1) Western Australia 308.4 (304.2–312.6) 304 (294.3–313.6) 331.5 (319.2–343.8) 357.6 (335–380.1) 436.5 (396.9–476.1) Tasmania d 380.6 (371.7–389.5) 392.9 (379.7–406.1) 417.2 (351–483.4) 450.2 (343.2–557.2) Northern Territory e 382 (356.7–407.4) 495.3 (453.1–537.5) 711 (656.1–766) Australia 327.5 (326.2–328.9) 351.8 (349.4–354.3) 369.3 (365.5–373.0) 404.0 (392.3–415.8) 512.4 (491.6–533.2) a The Remoteness Area classification is based 2006 census year . b Estimated Resident population for the year 2002–2006, rebased to 2006 Remoteness Areas . c There were no VR category areas in Victoria . d There were no MC category areas in Tasmania. Hobart is classified as an IR remoteness area . e There were no MC and IR category areas in the Northern Territory. Darwin is classified as an OR remoteness area . Selected chronic diseases are: diabetes (E10–E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19) . Discussion Australia has among the lowest mortality rates in the world with rates that have been consistently falling over the past century. The improvements in the early part of the century have been largely attributed to changes in living conditions while, more recently, improvements in health care have played an important role. However, the decline in mortality has not been shared equally by high and low socioeconomic groups. A mortality gap is also apparent based on the geographic variation of relative remoteness. The current study confirms that there is a marked gradient of increasing chronic disease mortality from major cities to remote and very remote areas of Australia. This pattern was observed in Australia as a whole, as well as within most states and territories, and for each of the six common chronic diseases. We estimate that, during the study period, there were a total of 11,900 deaths (3.3% of total) that may have been prevented if all remoteness categories had the same mortality rate as the MC category. The authors are not aware of previous studies that are directly comparable with this study. Similar studies have been limited to either individual states or territories or Australia as a whole and in most cases have excluded deaths for which locality was not available as a result of SLA boundary changes. We believe that this study provides nationally consistent results for chronic disease mortality differentials across remoteness categories and states and territories. There are a number of possible reasons for the observed association between relative remoteness and chronic disease mortality in Australia that may be related to either the context or the composition of the population. Firstly, shortages of healthcare professionals in rural and remote area present a serious challenge for access and continuity of healthcare delivery. In Australia, the distribution of health workforce remains geographically skewed, with concentrations of doctors and nurses in MC or IR areas much higher than R and VR areas. The shortage of general practitioners, medical specialists and dentists in remote and very remote areas will exacerbate the inequality in resource allocation in the Medical Benefit Scheme, Pharmaceutical Benefit Scheme and dental services. Second, many studies have identified social and economic factors as one explanation of inequality in health. Socioeconomic status (SES) is one of the most reliable descriptors of health disparities, with people from low SES experiencing poorer health and higher mortality. It is likely that lower SES people in more remote areas are less likely to benefit from universal health care than those at the higher end of SES. A third explanation is the higher proportion of Indigenous people in VR areas. Indigenous Australians have a significant gap in health outcomes and make up 47.4% of the total population in VR areas, compared to 15.2% in R areas, 5.8% in OR, 2.7% in IR and 1.2% in MC. There are a number of limitations to the study. The first is that the study depends on the reliability of death certificate data and subsequent cause of death coding. This limitation is offset by the strict coding conventions applied to the national deaths dataset and our use of multiple causes of death. Second, the study data is unlinked and numerator (mortality) and denominator (population) data came from different sources, for which there may be some inconsistency in SLA coding between different versions of ASGC available over the study period. This study minimised the misclassification, due to differences in SLA coding, by rebasing individual year datasets to 2006 SLAs. This approach may underestimate real differences at SLA, sex or age group level; however, this analysis was based on data aggregated to RA category, minimising the effect on the final results. A third limitation is that RA categories used in this study were based on the ABS 2006 Census, and calculated using road distance to the nearest service location, and extrapolated to Collection District (CD) and SLA geographic areas. This approach provides a useful summary measure but may underestimate the true extent of remoteness, particularly in larger SLAs. A fourth limitation is that Indigenous status was not included in the analysis, as it is not available in death data for all states and territories. This precluded the examination of the separate effect of Indigenous status on the differences in mortality across remoteness categories. A final limitation is that the study was unable to report mortality from 2007 onwards because of changes in the administrative responsibility for release of the national cause of death unit record file. Changes in health service funding, workforce and policies may have affected chronic disease mortality rates in recent years. Conclusion This study shows that more remote areas have a much higher chronic disease mortality rate than less remote areas, and that there is a gradient of increasing chronic disease mortality with increasing remoteness. This gradient also exists for individual chronic diseases and across most states and territories. Unlike similar studies, this study was able to accommodate the effect of SLA boundary changes on analysis by remoteness category. Acknowledgements This study was supported by the Department of Health, Northern Territory Government. The authors are grateful to the Australian Bureau of Statistics for providing the data. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Australian and New Zealand Journal of Public Health Wiley

Does relative remoteness affect chronic disease outcomes? Geographic variation in chronic disease mortality in Australia, 2002–2006

Download PDF
5 pages

Loading next page...
 
/lp/wiley/does-relative-remoteness-affect-chronic-disease-outcomes-geographic-6F2cuBGEFu

References (44)

Publisher
Wiley
Copyright
© 2013 Public Health Association of Australia
ISSN
1326-0200
eISSN
1753-6405
DOI
10.1111/1753-6405.12126
pmid
24812715
Publisher site
See Article on Publisher Site

Abstract

R ecent studies have reported differences in mortality outcomes between remote and urban populations, with a correlation between increasing remoteness and excess deaths observed in both Australia and overseas. Some researchers have described the scale of discrepancies while others have examined the cause of the differences. Studies in Australia have also highlighted the substantial mortality burden attributable to socioeconomic inequality and proposed that this may, in part, be the result of inequality in the individual benefits of healthcare. Poorer health care access, health risk factors and environmental health hazards have been associated with increasing remoteness. An additional factor that may also explain poorer mortality outcomes for remote areas is the higher concentration of Indigenous people who, as a population, have poorer health outcomes including mortality. Among the Indigenous population, the association of mortality and remoteness has varied, with two studies reporting lower mortality in very remote areas, while in Queensland, Indigenous death rates have recently been reported as higher in remote areas. Past studies investigating geographic remoteness and mortality have examined either Australia as a whole or selected states. In Australia, chronic diseases contribute significantly to premature mortality, but we know of no previous studies that have examined the six leading causes of chronic disease mortality across all states and territories by relative remoteness. Further, in this study we examined national death registration data and dealt with data discrepancies associated with geographic boundary changes that occurred during the study period. Historically, only underlying cause of death was available in death registration data although it is well documented that underlying cause of death alone underestimates the true burden of chronic disease mortality. More recent studies have used multiple causes of death, to more accurately estimate the impact of any one particular disease on mortality. This study examined the variation in multiple cause mortality rates from chronic diseases by relative remoteness for Australia as a whole, and within each state and territory. The study was approved by the Human Research Ethics Committee of the Northern Territory Department of Health and Menzies School of Health Research (HREC, 2011–1551). Methods Death data This study used Australian Bureau of Statistics (ABS) death registration data from 2002 to 2007 to identify deaths that occurred between 2002 and 2006 across all ages. The additional registration year (2007) allowed the inclusion of late registrations. Death registration data contains a deceased person's age; sex; usual residence coded to five‐digit Statistical Local Area (SLA); and disease diagnosis codes for underlying cause of death and associated causes of death. The coding, completed by ABS, used the International Classification of Diseases, 10 th Revision (ICD‐10). Six common chronic diseases were included: diabetes (E10‐E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19). Cases were selected using multiple causes of death (both underlying and associated causes) for the six chronic diseases. In ABS death registration data, usual residence addresses recorded in death certificates were assigned a five‐digit SLA code. These codes were based on the Australian Geographic Classification System (ASGC) SLA codes from the previous year. Historically, SLA geographic boundaries were modified annually to incorporate new local government council areas, expansion of suburban areas or realignment of town planning zones. Based on our analysis, it was estimated an average 6% of total SLAs in Australia were modified between 2002 and 2006. In this study, we mapped the 2002 to 2006 SLA level death data to 2006 SLAs using ABS population weighted correspondence files. Of the total 663,121 death records between 2002 and 2006, the number of deaths excluded from analysis was 2,411 (0.36%), which included deaths that occurred in the Off‐Shore Areas & Migratory SLAs and records with invalid or missing SLAs. Measure of Remoteness Area In 2001, ABS listed the Remoteness Area (RA) classification within the ASGC. The RA classification was based on the Accessibility/Remoteness Index of Australia (ARIA), initially developed by the Commonwealth Department of Health and Ageing (DoHA) and the National Centre for Social Applications of GIS (GISCA). The RA classification was derived using the road distance to the nearest urban centres in each of five categories based on population size. The RA classification defines Australian areas into the five categories: Major Cities (MC), Inner Regional (IR), Outer Regional (OR), Remote (R) and Very Remote (VR) areas. Population data The ABS estimated resident population (ERP) for the years from 2002 to 2006, rebased to 2006 SLAs, was obtained from ABS. The population data contains five‐year age group, sex, SLA code and states and territories identifier. The population data used in this study were consistent with death data after the previous adjustment, by ABS, from 2002 to 2005 SLA level ERPs to 2006 SLAs, using population weighted correspondence files. The SLA level death and population data were aggregated to RA categories by states and territories and Australia, using ABS population weighted correspondence file of 2006 RA to 2006 SLA. The study data, by RA categories, were further aggregated to three analysis groups: all persons; persons aged 35–64; and persons aged 65 years and above. Statistical analysis Age‐standardised mortality rates for the combined six chronic diseases for all persons, persons in the age groups 35–64 and 65 years and above, and rates for each individual chronic disease for all persons were calculated using direct standardisation to the 2001 Australian standard population. Chronic disease mortality rates for each RA category of states and territories and Australia were examined for the two age groups and all persons. A standardised rate ratio (SRR) for chronic disease deaths was used to measure the differences in mortality rate between capital cities of states and territories and very remote areas. All statistical analyses were conducted using STATA software, version 11. Results Overall chronic disease mortality During the period 2002–2006, there were 240,975 deaths with one of the six selected chronic diseases listed as the underlying cause and an additional 114,261 deaths with at least one of the chronic diseases listed as an associated cause of death. Of the total multiple causes of chronic disease deaths (355,236), 10.3% of deaths were in the age group 35–64 years and 89.3% deaths were in the age group 65 years and above, with the small balance (0.4%) among those aged less than 35 years. In the age group 35–64 years, the proportion of total deaths in each of the remoteness categories increased with increase in remoteness in both males and females. As a result, there was a converse distribution in the older age group with the proportion of total deaths by remoteness areas decreasing with increase in remoteness (Table ). Overall, at least one of the six chronic diseases was recorded as a cause of death for more than half (54%) of total deaths. Ischaemic heart disease was the leading cause of chronic disease death and was recorded as either the underlying or associated cause of death for half of total deaths (51%), followed by: stroke (28%); hypertension (18%); chronic obstructive pulmonary diseases (COPD, 17%); diabetes (16%); and chronic renal disease (13%). Number and proportion of chronic disease deaths for five remoteness categories a by age group and gender, 2002–2006, Australia. Male Female Person no % no % no % <35 years Major Cities Inner Region Outer Region Remote Very Remote Australia 581 190 102 41 66 978 0.5 0.5 0.5 1.5 4.8 0.5 367 127 73 26 19 604 0.3 0.3 0.4 1.4 2.0 0.3 948 317 175 67 85 1,582 0.4 0.4 0.5 1.5 3.7 0.4 35–64 years Major Cities Inner Region Outer Region Remote Very Remote Australia 15,303 5,629 3,291 704 573 25,500 13.4 13.5 16.5 26.5 41.4 14.2 6,547 2,507 1,362 328 361 11,105 5.6 6.4 8.2 17.4 38.2 6.3 21,850 8,136 4,653 1,032 934 36,605 9.5 10.1 12.8 22.7 40.1 10.3 65 years & above Major Cities Inner Region Outer Region Remote Very Remote Australia 98,410 35,991 16,541 1,915 744 153,602 86.1 86.1 83.0 72.0 53.8 85.3 109,960 36,285 15,103 1,526 564 163,447 94.1 93.2 91.3 81.2 59.7 93.3 208,370 72,276 31,644 3,441 1,308 317,049 90.1 89.5 86.8 75.8 56.2 89.3 All persons Major Cities Inner Region Outer Region Remote Very Remote Australia 114,294 41,810 19,934 2,660 1,383 180,080 100.0 100.0 100.0 100.0 100.0 100.0 116,874 38,919 16,538 1,880 944 175,156 100.0 100.0 100.0 100.0 100.0 100.0 231,168 80,729 36,472 4,540 2,327 355,236 100.0 100.0 100.0 100.0 100.0 100.0 a The Remoteness Area classification is based 2006 census year . Selected chronic diseases are: diabetes (E10–E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19) . Chronic disease mortality rates by Remoteness Area (RA) at national level Table shows age‐standardised chronic disease mortality rates for age groups: 35–64 years; 65 years and above; and all persons. The rates were lowest among those living in MC and consistently increased toward VR (Table ). This pattern was seen in both males and females and both age groups. A standardised rate ratio was used to compare mortality rates in different remoteness categories with the rate in the MC category. The SRR to MC in the age group 35–64 years is much higher than age group 65 and above. In the age group 35–64, the mortality rate in VR is 4.2 times higher than MC, female rates much higher (6.2 times) than males rates (3.2 times). In the age group 65 and above, the SRR is moderately increased with remoteness; in VR the mortality rate among males is 1.2 times higher than MC and 1.3 times higher among females. Age standardised chronic disease mortality rates (and 95%CI) for five remoteness categories a by age group and gender, 2002–2006, Australia. Male Female Person Rate 95% CI Rate 95% CI Rate 95% CI Deaths per 100,000 population b 35–64 years Major Cities Inner Region Outer Region Remote Very Remote Australia 113.7 129.6 151.8 210.0 368.3 124.7 (111.9–115.5) (126.2–133.0) (146.6–157.0) (194.6–225.6) (338.2–398.5) (123.1–126.2) 47.9 57.9 67.9 117.5 294.5 54.4 (46.7–49.1) (55.6–60.2) (64.3–71.5) (104.8–130.2) (264.1–324.9) (53.5–55.5) 80.5 93.9 111.5 168.1 335.5 89.6 (79.4–81.6) (91.8–95.9) (108.3–114.7) (157.8–178.3) (314.0–357.1) (88.7–90.5) 65 years & above Major Cities Inner Region Outer Region Remote Very Remote Australia 2,721.0 2,892.0 3,011.0 3,090.8 3,258.6 2,796.8 (2,704.0–2738.0) (2,862.2–2922.0) (2,965.1–3056.9) (2,952.4–3229.2) (3,025.0–3492.7) (2,782.8–2810.8) 2,078.0 2,186.2 2,216.2 2,271.0 2,670.0 2,118.2 (2,065.7–2090.3) (2,163.7–2208.7) (2,180.8–2251.5) (2,157.1–2385.0) (2,449.5–2890.4) (2,107.9–2128.4) 2,353.4 2,502.4 2,585.4 2,672.2 2,984.4 2,415 (2,343.3–2363.5) (2,484.1–2520.6) (2,556.9–2613.8) (2,583.0–2761.5) (2,822.6–3146.1) (2,406.6–2423.4) All persons Major Cities Inner Region Outer Region Remote Very Remote Australia 386.7 414.9 438.4 473.4 563.7 400.8 (384.5–388.9) (410.9–418.9) (432.3–444.5) (455.4–491.4) (534.0–593.4) (398.9–402.7) 280.2 297.9 305.9 333.5 452.6 287.9 (278.6–281.8) (295.0–300.9) (301.2–310.6) (318.4–348.5) (423.7–481.5) (286.6–289.2) 327.5 351.8 369.3 404.0 512.4 339.0 (326.2–328.9) (349.4–354.3) (365.5–373.0) (392.3–415.8) (491.6–533.2) (337.9–340.1) a The Remoteness Area classification is based 2006 census year . b Estimated Resident population for the year 2002–2006, rebased to 2006 Remoteness Areas . Selected chronic diseases are: diabetes (E10–E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19) . A similar trend was also observed for individual chronic diseases for all persons, with MC having the lowest mortality rate increasing towards VR (Table ). There was a minor reversal in the point estimates for stroke between IR and OR remoteness categories. Age standardised mortality rates (and 95% CI) for six chronic diseases by remoteness category, a 2002–2006, Australia. Chronic diseases Major Cities (MC) Inner Region (IR) Outer Regional (OR) Remote (R) Very Remote (VR) Australia (AUS) Deaths per 100,000 population b Diabetes 51.6 (51.1–52.1) 55.4 (54.4–56.3) 64.8 (63.3–66.4) 84.1 (78.9–89.4) 139.7 (129.1–150.3) 54.5 (54.1–54.9) Ischaemic heart disease 169.6 (168.6–170.6) 180.1 (178.4–181.8) 183.8 (181.1–186.4) 190.7 (182.6–198.7) 222.3 (208.8–235.7) 174 (173.2–174.8) Stroke 93.3 (92.5–94.0) 95.2 (94.0–96.5) 93.9 (92.0–95.8) 96.5 (90.7–102.4) 106.1 (96.2–116.0) 94 (93.4–94.6) Hypertension 59.7 (59.1–60.3) 60.8 (59.8–61.8) 67.0 (65.4–68.7) 78.2 (73.0–83.4) 96.7 (87.5–105.8) 61.1 (60.6–61.6) Chronic obstructive pulmonary disease 53.6 (53.0–54.1) 62.8 (61.8–63.8) 71.0 (69.4–72.7) 81.3 (76.0–86.5) 113.5 (103.5–123.6) 57.9 (57.4–58.4) Renal 44.3 (43.8–44.8) 45.0 (44.2–45.9) 48.0 (46.7–49.4) 60.3 (55.8–64.9) 92.5 (83.7–101.3) 46.6 (46.2–47.0) a The Remoteness Area classification is based 2006 census year . b Estimated Resident population for the year 2002–2006, rebased to 2006 Remoteness Areas . Chronic disease mortality rates by Remoteness Area (RA) for states and territories Age‐standardised mortality rates for chronic diseases also varied across RAs within state and territories; again there was a general increase in mortality rates with increasing remoteness, with the exception of Victoria (Table ). In Victoria, the point estimate for the R mortality rate was lower than either the IR or OR categories, though there was no statistical difference between the three categories. The greatest increase in mortality rate with increased remoteness was in the NT, where the VR mortality rate was 1.9 times higher than the Darwin area – classified as OR in remoteness categories. Age standardised chronic disease mortality rates (and 95% CI) for States and Territories by remoteness category, a 2002–2006, Australia. Major Cities (MC) Inner Regional (IR) Outer Regional (OR) Remote (R) Very Remote (VR) Deaths per 100,000 population b New South Wales & Australian Capital Territory 336.3 (334.1–338.6) 351.9 (347.9–356) 366.3 (359.2–373.5) 448.9 (415.9–481.9) 490.3 (390–590.5) Victoria c 323.5 (321–326.1) 360.6 (355.7–365.4) 369.6 (360.3–378.8) 349.3 (286.8–411.7) Queensland 328.4 (325–331.7) 352.1 (346.6–357.6) 367.7 (360–375.3) 409.2 (385.6–432.8) 479.5 (446.6–512.5) South Australia 323.9 (319.7–328.1) 311.4 (300.8–322) 379.8 (368.3–391.3) 360.6 (337.1–384.1) 451.5 (395–508.1) Western Australia 308.4 (304.2–312.6) 304 (294.3–313.6) 331.5 (319.2–343.8) 357.6 (335–380.1) 436.5 (396.9–476.1) Tasmania d 380.6 (371.7–389.5) 392.9 (379.7–406.1) 417.2 (351–483.4) 450.2 (343.2–557.2) Northern Territory e 382 (356.7–407.4) 495.3 (453.1–537.5) 711 (656.1–766) Australia 327.5 (326.2–328.9) 351.8 (349.4–354.3) 369.3 (365.5–373.0) 404.0 (392.3–415.8) 512.4 (491.6–533.2) a The Remoteness Area classification is based 2006 census year . b Estimated Resident population for the year 2002–2006, rebased to 2006 Remoteness Areas . c There were no VR category areas in Victoria . d There were no MC category areas in Tasmania. Hobart is classified as an IR remoteness area . e There were no MC and IR category areas in the Northern Territory. Darwin is classified as an OR remoteness area . Selected chronic diseases are: diabetes (E10–E14); ischaemic heart disease (I20–I25); stroke (I60–I69); hypertension (I10); chronic obstructive pulmonary disease (J40–45, J47); and renal disease (N18–N19) . Discussion Australia has among the lowest mortality rates in the world with rates that have been consistently falling over the past century. The improvements in the early part of the century have been largely attributed to changes in living conditions while, more recently, improvements in health care have played an important role. However, the decline in mortality has not been shared equally by high and low socioeconomic groups. A mortality gap is also apparent based on the geographic variation of relative remoteness. The current study confirms that there is a marked gradient of increasing chronic disease mortality from major cities to remote and very remote areas of Australia. This pattern was observed in Australia as a whole, as well as within most states and territories, and for each of the six common chronic diseases. We estimate that, during the study period, there were a total of 11,900 deaths (3.3% of total) that may have been prevented if all remoteness categories had the same mortality rate as the MC category. The authors are not aware of previous studies that are directly comparable with this study. Similar studies have been limited to either individual states or territories or Australia as a whole and in most cases have excluded deaths for which locality was not available as a result of SLA boundary changes. We believe that this study provides nationally consistent results for chronic disease mortality differentials across remoteness categories and states and territories. There are a number of possible reasons for the observed association between relative remoteness and chronic disease mortality in Australia that may be related to either the context or the composition of the population. Firstly, shortages of healthcare professionals in rural and remote area present a serious challenge for access and continuity of healthcare delivery. In Australia, the distribution of health workforce remains geographically skewed, with concentrations of doctors and nurses in MC or IR areas much higher than R and VR areas. The shortage of general practitioners, medical specialists and dentists in remote and very remote areas will exacerbate the inequality in resource allocation in the Medical Benefit Scheme, Pharmaceutical Benefit Scheme and dental services. Second, many studies have identified social and economic factors as one explanation of inequality in health. Socioeconomic status (SES) is one of the most reliable descriptors of health disparities, with people from low SES experiencing poorer health and higher mortality. It is likely that lower SES people in more remote areas are less likely to benefit from universal health care than those at the higher end of SES. A third explanation is the higher proportion of Indigenous people in VR areas. Indigenous Australians have a significant gap in health outcomes and make up 47.4% of the total population in VR areas, compared to 15.2% in R areas, 5.8% in OR, 2.7% in IR and 1.2% in MC. There are a number of limitations to the study. The first is that the study depends on the reliability of death certificate data and subsequent cause of death coding. This limitation is offset by the strict coding conventions applied to the national deaths dataset and our use of multiple causes of death. Second, the study data is unlinked and numerator (mortality) and denominator (population) data came from different sources, for which there may be some inconsistency in SLA coding between different versions of ASGC available over the study period. This study minimised the misclassification, due to differences in SLA coding, by rebasing individual year datasets to 2006 SLAs. This approach may underestimate real differences at SLA, sex or age group level; however, this analysis was based on data aggregated to RA category, minimising the effect on the final results. A third limitation is that RA categories used in this study were based on the ABS 2006 Census, and calculated using road distance to the nearest service location, and extrapolated to Collection District (CD) and SLA geographic areas. This approach provides a useful summary measure but may underestimate the true extent of remoteness, particularly in larger SLAs. A fourth limitation is that Indigenous status was not included in the analysis, as it is not available in death data for all states and territories. This precluded the examination of the separate effect of Indigenous status on the differences in mortality across remoteness categories. A final limitation is that the study was unable to report mortality from 2007 onwards because of changes in the administrative responsibility for release of the national cause of death unit record file. Changes in health service funding, workforce and policies may have affected chronic disease mortality rates in recent years. Conclusion This study shows that more remote areas have a much higher chronic disease mortality rate than less remote areas, and that there is a gradient of increasing chronic disease mortality with increasing remoteness. This gradient also exists for individual chronic diseases and across most states and territories. Unlike similar studies, this study was able to accommodate the effect of SLA boundary changes on analysis by remoteness category. Acknowledgements This study was supported by the Department of Health, Northern Territory Government. The authors are grateful to the Australian Bureau of Statistics for providing the data.

Journal

Australian and New Zealand Journal of Public Health – Wiley

Published: Apr 1, 2014

There are no references for this article.