Methods

Mapping tables between ICD-9-CM and ICD-10-AM disease codes, produced by the National Centre for Classification in Health, were used extensively to obtain the most appropriate match for individual codes between the two classification systems. The ICD-10-AM and ICD-9-CM codes used for each indicator are included in the disease and procedure codes section of the appendix.

2.5 NSW Midwives Data Collection

The New South Wales Midwives Data Collection (MDC) is a population-based collection covering all births in NSW public and private hospitals, as well as home births. It does not receive notifications of interstate births where the mother is resident in NSW.

The data collection has operated continuously since 1990 and covers: up to 2005, all births in NSW of at least 400 grams birth weight or 20 weeks gestation; and for 2006, stillbirths of at least 400 grams birthweight or 20 weeks gestation and all live births. The information is recorded by either the midwife or medical practitioner and includes demographic, medical and obstetric information on the mother and information on the labour, delivery and condition of the infant.

The collection has undergone three revisions over the years. Data on antenatal visits have been collected since 1994. The MDC database is compiled in the Information Management and Support Branch of the NSW Department of Health.

For this report, the MDC was accessed via HOIST. Data are presented for calendar years.

Refer to 2.3 for information on perinatal deaths derived from the Australian Bureau of Statistics.

2.6 NSW Birth Defects Register

The NSW Birth Defects Register (BDR) was established in 1990. Legislation to mandate the notification of birth defects recognised at up to one year of age has been in effect since 1 January 1998. Prior to that, the BDR operated on a voluntary reporting basis.

The NSW BDR collects clinical and demographic data about pregnancies affected by a structural or chromosomal defect.

For this report, the BDR was accessed via HOIST. Data are presented for calendar years.

2.7 NSW Population Health Survey

Since 2002, the NSW Department of Health, in conjunction with the area health services, has conducted the New South Wales Population Health Survey, an ongoing survey of the health of people in NSW using computer-assisted telephone interviewing (CATI). The main aims of the NSW Population Health Survey are to provide detailed information on the health of the people of NSW, and to support the planning, implementation, and evaluation of health services and programs in NSW.

The target population for the NSW Population Health Survey is all NSW residents living in households with private telephones. The target sample comprises approximately 1,500 people in each area health service (total sample of 12,000). Households are contacted using list-assisted random digit dialling. Up to seven calls are made to establish initial contact with a household and up to five calls are made to contact a selected respondent. One person from the household is randomly selected for inclusion in the survey. Carers or parents of children aged 0-15 years are interviewed on their behalf.

Respondents are asked questions from modules on demographics, health behaviours, health status, and access to and satisfaction with health services. Additional question modules are added periodically and are reported less frequently. Trained telephone interviewers carry out the interviews. Most interviews are conducted in English but the survey is also conducted in five other languages: Arabic, Chinese, Greek, Italian and Vietnamese. The sample is weighted to adjust for differences in the probabilities of selection among subjects, and for differences between the age and sex structure of the sample and Australian Bureau of Statistics mid-year population estimates for NSW. This enables calculation of prevalence estimates for the state population rather than for the respondents selected.

In 2007, 16,046 interviews were conducted, with 13,178 adults aged 16 years or over. At least 1550 persons were interviewed in each Area Health Service. The overall response rate was 63.6 per cent (completed interviews divided by completed interviews plus refusals).

Child (0-15 years) component was included in the NSW Population Health Survey in 2003. The reporting plan for the continuous survey includes an annual report on adult health for the whole state, annual reports on adult health for selected indicators by area health service and a biennial report on child health for the whole state. The first two reports on child health from the continuous survey reported data from 2003 and 2004 and from 2005 and 2006. Triennial reports on adult health for the divisions of general practice are also available: the first two reports cover the combined years 2002, 2003 and 2004 ('2004' report published in 2005) and 2005, 2006 and 2007 ('2007' report published in 2008). Also available are reports on adult Aboriginal health, adult health by country of birth (both reports covering years 2002-2005) and reports on health of young adults (16-24 years), adults 45 years and over and older people (65 years and over) all covering years 1997-2007.

For this report, data from the NSW Health Survey for special populations were accessed via published reports or creation of separate datasets for the CHO Report by members of the NSW Health Survey Program.

2.7.1 NSW Health Surveys 1997 and 1998, NSW Child Health Survey 2001, NSW Older People's Health Survey 1999

Prior to the introduction of the continuous survey in 2002, the Centre for Epidemiology and Research conducted adult health surveys in 1997 and 1998, an older people's health survey in 1999, and a child health survey in 2001.

For this report, data from these Surveys were accessed via HOIST.

2.8 ABS National Health Survey

The National Health Surveys (NHS) conducted by the ABS collect information on illness and injury, health care use, and health risk factors. Data from the 1989-90, 1995, 2001 and 2004-05 National Health Surveys were assessed and data from 2004-05 (published in 2006) are presented in this report. Data were accessed via HOIST, were obtained as special tabulations from the ABS, or were from published reports.

2.9 ABS National Survey of Mental Health and Wellbeing

The National Survey of Mental Health and Wellbeing was conducted by the ABS from August to December 2007. The full results were published in 2008 (ABS, Cat no 4326.0).

The survey collected information from approximately 8,800 Australians aged 16-85 years. Reports provide information on the prevalence of selected lifetime and 12-month mental disorders by three major disorder groups: Anxiety disorders (e.g. Social Phobia), Affective disorders (e.g. Depression) and Substance Use disorders (e.g. Alcohol Harmful Use). Also provided is information on the level of impairment, the health services used for mental health problems, physical conditions, social networks and caregiving, as well as demographic and socioeconomic characteristics.

The previous survey, the National Survey of Adult Mental Health and Wellbeing was conducted by the ABS in 1997-1998 (ABS, Cat no 4326.0). Various reports from this Survey, published by the ABS between 1998 to 2001, were consulted, as was information on mental health available from the National Health Surveys conducted by the ABS in 2001 (ABS, Cat no 4811.0) and 2004-05 (ABS, Cat no 4364). Data were accessed via HOIST, were obtained as special tabulations from the ABS, or were from published reports.

2.10 School surveys

2.10.1 ASSAD/NSW School Student Health Behaviours Survey

The NSW Department of Health and The Cancer Council NSW have carried out surveys of the health of secondary school students since 1984 as part of the triennial Australian School Students' Alcohol and Drug (ASSAD) survey.

In its earlier years, the ASSAD Survey questions targeted drug and alcohol use. The topics covered by the Survey have gradually extended to include other issues that are important to the health of adolescents. In 2005 the NSW School Student Health Behaviours Survey included questions on physical activity and injuries, sun protection behaviours, eating behaviours, and mental health and wellbeing, in addition to smoking, alcohol and other drug use.

ASSAD/NSW School Student Health Behaviours Survey data have been collected in a consistent way over time and are the most reliable current source of information on trends in secondary school students' health.

2.10.2 The NSW Schools Physical Activity and Nutrition Survey

About 5,400 students from Kindergarten and Years 2, 4, 6, 8 and 10 were studied in early 2004. They were drawn from all types of schools in NSW, from both city and country, and were representative of the NSW population. Researchers gathered basic demographic data and measured the height, weight and waists of all students. Availability of physical measurements makes this survey especially important. Further details can be found in three reports from the Survey: Full report, Summary report and Short report (NSW COO, 2004; Booth et al., 2004), published by the NSW Centre for Overweight and Obesity and the NSW Department of Health.

2.11 NSW Central Cancer Registry data

The NSW Central Cancer Registry (CCR) was established by the NSW Department of Health in 1971 under the NSW Public Health Act. It was administered by the Cancer Council NSW, under contract, from 1986 until June 2004. The Registry has been managed by the Cancer Institute NSW since June 2004.

Notification of newly diagnosed cases of, and deaths due to, malignant neoplasm is a statutory requirement for public and private hospitals, pathology laboratories, day procedure centres, departments of radiation oncology, outpatient departments and nursing homes. Death certificates are provided by the NSW Registry of Births, Deaths and Marriages. Data collected on notification forms include identifying and demographic information, brief medical details describing the cancer and a record of at least one episode of care from each notifier. These data are supplemented by pathology reports and death certificates. For breast cancer and cutaneous melanoma additional prognostic factors are coded from pathology reports and in situ lesions are registered. Notification has traditionally been via a printed notification form, although in recent years electronic notification by hospitals (but not pathology laboratories) has been introduced.

Multiple primary cancers in the same person are counted according to rules set out by the International Association of Cancer Registries. The term 'cases' refers to ether newly diagnosed cancers or deaths. Cases are the basis for reporting in this report and in the reports by the NSW Cancer Institute. Only cancer prevalence is reported as a number of individuals living with cancer as well as a number of cases of cancers in living persons in this report. Deaths of individuals from cancer refer to a one primary cancer listed as the cause of death (an underlying casue of death).

Incident cases and deaths registered before July 1999 were classified according to the 9th revision of the International Classification of Diseases (ICD-9, World Health Organization, 1977). Cases registered from July 1999 onwards have been classified according to the 2nd edition of the International Classification of Diseases for Oncology (ICD-O-2, World Health Organization, 1990). ICD-O-2 codes were mapped to ICD-10-AM codes by the Registry. This report uses ICD-9 and ICD-10 codes. Leading cancers by type are analysed using ICD-O-3 and mesothelioma trend uses ICD-O-3 and its predecessors (that is, values from the 'topotab' variable, Cancer Library, HOIST).

The cancers required to be notified fall within the following ICD-10-AM codes: C00-C76 and C80-C96. Codes C77-C79 refer to secondary malignant neoplasms from primary sites which have been already a subject of notification. Notification of basal and squamous cell carcinoma of the skin is not required (C44 topography code applies where morphology code is between M805-M811). Squamous cell carcionomas of the lip, vulva, penis, scrotum and anus are registered when notified but data are not routinely reported by the CCR.

For this report, cancer incidence data were supplied by the NSW Central Cancer Registry and accessed via HOIST. The cancer deaths data presented here are from the Australian Bureau of Statistics mortality collection. The ABS death data were used to maintain consistency with other chapters of the report. However, generally the CCR figures for cancer are more reliable than the ABS cancer data, because the CCR data are verified against cases' histopathological reports while the ABS data are sourced only from the information on death certificates.

2.12 NSW Notifiable Diseases Database

The NSW Notifiable diseases database (NDD), formerly called the NSW Infectious Diseases Surveillance system, is a networked database used by eight public health units (PHUs) located across NSW to register communicable disease notifications. Under authority of the NSW Public Health Act 1991, the NSW Health Department receives notifications of communicable disease via PHUs from general practitioners, hospitals, and pathology laboratories. Data are transferred weekly from PHUs to the Department, for compilation of statewide data. The Department transfers a smaller dataset to the Communicable Diseases Network of Australia and New Zealand (maintained by the Commonwealth Department of Health and Ageing). Annual statewide data becomes available in April of the following year after processing aiming to remove duplicate records and errors.

For this report, the NDD collection was accessed via HOIST.

2.13 Australian Childhood Immunisation Register

The Australian Childhood Immunisation Register (ACIR), which is managed by the Health Insurance Commission and commenced operation on 1 January 1996, is a register of the immunisation status of all children less than seven years of age. Broadly, the functions of the ACIR are to administer a national recall-reminder service to parents, to provide immunisation status information to parents and providers and to provide immunisation coverage data. The ACIR collects immunisation information from immunisation providers and administers a Commonwealth-State cost-shared payments system to providers for reporting information.

ACIR supplies NSW Health with monthly coverage data that identifies children 'overdue' for immunisation, which are forwarded to Public Health Units for follow up, and quarterly coverage data by local government area. These latter data form the basis for the information presented in this report.

2.14 NSW Inmate Health Surveys 1996, 2001 and 2008

The overall aims of the Inmate Health Surveys are to describe the health of adult inmates, to identify factors associated with poor health, to develop indicators allowing comparisons to be made with the health of the general population, and to develop health goals and targets for the inmate population based on the findings. Special emphasis was placed on determining the health of Aboriginal and elderly inmates.

The methodology of surveys is similar to ensure consistency across the surveys. The design represents a cross-sectional random sample of inmates stratified by sex, age and Aboriginality. A sample includes approximately 10% of male and more than 30% of female inmates in full-time custody. All correctional facilities in NSW are included in the surveys.

Aboriginal people are over-represented in the correctional system, comprising about 15% and 25% of male and female prisoners compared with approximately 1% in the general community. Given this overrepresentation, and variations in health status between Aboriginal and non-Aboriginal Australians, survey samples are stratified by Aboriginality. The samples are also stratified into three age groups: under 25 years, 25-40 years, and over 40 years. The stratification ensures that there are sufficient numbers of both Aboriginal and non-Aboriginal inmates to enable the health status of each sub-group could be described separately. In the 2001 survey the overall response rate was 85% (700 males and 154 females). The response rates were 84% for women and 85% for men; and 83% for Aboriginal people compared with 85% for non-Aboriginal people.

2.15 Emergency Department Data Collection

The NSW Emergency Department Data Collection is a database of information collected from approximately one-third of NSW Emergency Departments (EDs). It represents approximately two-thirds of all NSW emergency patients. Analyses included in this report are based on a provisional diagnosis assigned by staff when a patient presents to the Emergency Department.

Data collected by the EDs are mainly used for ED performance monitoring by ED staff. It is not possible to ensure that all variables are accurate and that they conform to current ED data standards before data from individual EDs are made available on HOIST. Additionally, since 2007 many EDs have changed to a new software and different system for coding provisional diagnosis, which is based on SNOMED classification. As previously, provisional diagnoses make extensive use of signs and symptoms and their groupings. Mapping of selected ED observations into discrete morbid entities, which are recognised and can be coded within the ICD classification, has been maintained by HOIST staff before ED data are placed on HOIST.

In view of these limitations, available ED data cannot be used extensively in the CHO Report as publication carries a risk of 'over-interpretation'. The ED data that are published in the report are accompanied by notes urging particular caution and attention to data limitations.

For this report, data from the Emergency Department Data Collection were accessed via HOIST.

3. Statistical methods

3.1 Crude death rates

Crude rates represent an estimate of the proportion of a population that experiences an outcome during a specified period. The crude rate is calculated by dividing the number people with an outcome in a specified period by the number at risk during that period (typically per year). It does not take into account the age structure of the population studied, and can be misleading when long-term trends are examined - or geographic areas are compared - because age structures of populations may vary over time or among areas.

3.2 Age-adjusted rates

Age-adjustment of rates used direct age-standardisation. This method adjusts for effects of differences in the age composition of populations across time or geographic regions. The directly age-standardised rate is the weighted sum of age-specific (five-year age group) rates, where the weighting factor is the corresponding age-specific standard population. For this report, the Australian estimated residential population (persons) as at 30 June 2001 was used as the standard population (this is given in Table 1). The same population was used for males and females to allow valid comparison of age-standardised rates between the sexes.

Ninety-five per cent confidence limits around the directly standardised rates were calculated using the method described by Dobson et al. (1991). This method gives more accurate confidence limits than the usual normal approximation for rarer conditions. Where the number of events is larger, the limits are equivalent to those calculated in the conventional fashion (Armitage, Berry and Matthews, 2002).

3.3 Life expectancy at birth

Life expectancy at birth is an estimate of the average length of time (in years) a person can expect to live, assuming that the currently prevailing rates of death for each age group will remain the same for the lifetime of that person. In fact, death rates will almost certainly change over the lifetime of a person born now, owing to changes in social and economic conditions, changes in lifestyle, advances in health care, and possibly the emergence of new diseases. However, because no-one knows what the death rates for each age group and sex will be in the future, the usual practice is to use the current rates of death to calculate life expectancy.

For this report, estimates and confidence intervals for life expectancy were calculated using abridged current life tables based on five-year age groups, except for the first five years of life, which were split into two age groups 0-<1 years and 1-4 years. The methods used are described in detail by Chiang (1984).

A life table is a statistical method used to represent mortality of a population. In its simplest form, a life table is genrated from age-specific death rates and the resulting values are used to measure mortality, survivorship and life expectancy.

Australia's life expectancy is ranked among the highest in the world. Methods used to calculate life expectancy differ between countries resulting in slight differences in the final figures reported by countries. These differences are not considerable and the group of countries that leads has been stable for a number of years, however their ranking may change somewhat from year to year. One of the differences between methods is the use of complete or abridged life tables, resulting in different age intervals used in the calculations. For example, the ABS publishes complete life tables for the Australian population (that is the tables contain data by single years of age) and abridged tables (for five-year age groups) for the Aboriginal and Torres Strait population. Another difference between methods stems from using one year or several combined years of death data as a smoothing technique designed to reduce the effect of year-to-year data variations. The ABS uses three combined years of death data. Other differences involve excluding, or otherwise, of residents who are physically not present at the time of count and actuarially graduating the tables. Both these adjustements are applied by the ABS (ABS, 2008).

The World Health Organization (WHO) calculates and reports estimates for all countries based on the information about population, births and deaths supplied by the countries. In this instance, although the method used is the same for all countries, the results are limited by the accuracy and timelines of information supplied by the countries.

Constructing a life table requires data on deaths and total population. Births data are also needed for adjusting the size of the total population. Because of uncertainty about the estimates of these components for Aboriginal peoples, indirect experimental menthods were previously used by the ABS to calculate life expectancies for the Aborriginal population in Australia. The ABS recently reviewed available methods with the view of using the one most appropriate for the circumstances of increasing self-identification. Three different methods that could be used to estimate life expectancy and which are discussed in ABS publications (ABS&AIHW, 2008) gave different results ranging from 20 years lower than the life expectancy estimates derived for all Australian males and females to 17 years and 13 years lower. The ABS has concluded that the previously used (indirect methods) are inappropriate for Australia .The estimates quoted in this report should only be used as indicative summary measures and are no longer regarded as reliable.

3.4 Life expectancy at age 65

The average number of additional years a person who has reached the age of 65 would expect to live if current mortality trends continue to apply is based on the age-specific death rates for a given year. This measure assumes that death rates will remain constant for the next 20 to 30 years, a much more conservative assumption than the one used to calculate life expectancy at birth. For this report, life expectancy was calculated using abridged current life tables based on five-year age groups.

3.5 Analysis of the NSW Population Health Survey

Respondents are asked questions from modules on demographics, health behaviours, health status, and access to and satisfaction with health services. Additional question modules are added periodically and are reported less frequently. Trained telephone interviewers carry out the interviews. Most interviews are conducted in English but the survey is also conducted in five other languages: Arabic, Chinese, Greek, Italian and Vietnamese. The sample is weighted to adjust for differences in the probabilities of selection among subjects, and for differences between the age and sex structure of the sample and Australian Bureau of Statistics mid-year population estimates for New South Wales. This enables calculation of prevalence estimates for the state population rather than for the respondents selected.

The 'Surveymeans' procedure in SAS for Windows Version 9.1 was used to calculate point estimates and 95% confidence intervals. This procedure uses the Taylor expansion method to estimate sampling errors of estimators based on a stratified random sample (SAS, 2005).

3.6 Graphical presentation

Figure 1 below demonstrates the method used for graphical presentation of point estimates and associated 95% confidence intervals. It shows age-standardised cardiovascular disease deaths for the years 2002 to 2006 combined for each of the NSW health areas, three ARIA+ geographic categories and NSW as a whole. The point estimates of standardised rates are indicated by horizontal bars, and 95% confidence intervals by horizontal lines. Point estimates of standardised rates are also listed beside each graph.

Figure 1. Sample graph demonstrating point estimates and 95% confidence intervals.

3.7 Analysis of indicators by local government area

The local government areas (LGA) are the smallest level at which data are analysed in this report. 'Statistical smoothing' methods used to stabilise the resulting small area estimates, are discussed in detail in the following sections.

There were 153 LGAs in NSW in 2008, using the 2006 ASGC boundaries. Due to the spatial isolation of Lord Howe Island, both cases and population on Lord Howe Island SLA (population 364 persons) were omitted from analysis and only Unincorporated Far West SLA (population 756 persons) was included as Unincorporated NSW LGA. The 153 LGAs ranged in population from 756 to almost 280,000 (based on population estimates as at June 2006). The distribution of these populations across the local government areas is shown in the Demography chapter.

There were 13 LGAs with total populations less than 3000, and of these five had populations less than 2000 (based on population estimates as at June 2006). The areas with the smallest populations are particularly vulnerable to the effect that chance events have on the rate or ratio.

3.7.1 Smoothing of estimates for small areas using statistical smoothing (Bayesian smoothing methods)

Mapping cases or rates of events of interest, such as rates of deaths, cases of a disease, or rates of smoking, can be very informative in understanding the geographical distribution of the events. However, low numbers and rates can occur if the event is rare or if the areas studied have small populations. If numbers or rates are low, then they will also be very variable, since chance events will have an undue effect on the total number. Consequently, estimates of numbers or rates may be too changeable to be reliable for most purposes. Occasionally, there may not be any cases of interest at all in an area, and the estimated rate for that area would be zero.

More reliable estimates of numbers and rates can be obtained by extending the length of time within which the cases are counted, or by increasing the size of the areas considered, but both these methods may contradict the purpose of the study and undermine the usefulness of the data. Another option is to apply statistical smoothing methods to calculate more reliable estimates with data collected in a shorter, more up-to-date period Rates can be estimated even when there were no cases in an area in the relevant period of time.

Statistical smoothing methods are used to improve the estimates for individual areas by including information on events in adjacent areas which are expected to be similar, and overall variability between all areas. Smoothing has the greatest effect for areas where the number of cases is the lowest.

In this report, Bayesian smoothing was used to adjust raw estimates by taking into account information from adjacent areas (local or spatial variability) and from the whole state (global or non-spatial variability).

For population-based indicators such as rates of hospitalisation and rates of death, Bayesian smoothing was performed using the convolution or Besag, York and Mollie (BYM) model (Lawson et al, 2003). This model is widely used for disease mapping. The smoothed estimates calculated are the age-standardised relative risk for each area compared to NSW. That is, the standardised incidence ratio (SIR) for hospitalisations and the standardised mortality ratio (SMR) for deaths. All models were fit using WinBUGS 1.4.3 (WinBUGS, 2007). Further details are provided in sections 3.7.1.1 and 3.7.1.2.

For indicators such as smoking in pregnancy and attendance at antenatal care, which are based on binary outcomes, smoothing was obtained by modelling the data using a binomial distribution with a logit link function. Smoothed proportions incorporate both local and global information, but are not age standardised. The details are further discussed in section 3.7.1.2.

The results of the Bayesian smoothing were used to determine whether the results obtained from individual areas are significantly different from the estimate of the average for all areas. Smoothed estimates are displayed on choropleth maps for all indicators. The intensity of the colour of a LGA increases as the ratio increases, and the same scale is used for all maps. The level of significance and the direction of difference from the state average is shown using '+ 'and '-' signs. One plus sign means that the smoothed estimate for a LGA is significantly greater than the state average at the 5% level of significance and two plus signs mean that the estimate is significantly greater at the 1% level of significance. Conversely, one minus sign means that the smoothed estimate for a LGA is significantly lower than the state average at the 5% level of significance and two minus signs refer to the 1% level of significance. If an area does not differ from the state, then no symbol is shown. All maps were produced using SAS V9.1.3 (SAS, 2005)

The success of the Bayesian smoothing method depends largely on the degree of similarity between areas that are used in the calculations. In the case of Local Government Areas in NSW, similarity is very high and the method works well.

3.7.1.1 Smoothing of estimates for population-based indicators

The smoothing used for the population-based indicators is obtained by applying the convolution or Besag, York and Mollie (BYM) model (Lawson et al, 2003). This model incorporates both spatially correlated and uncorrelated variation. It accounts for variability across the entire state (uncorrelated or global variation) as well as variability amongst the local government areas immediately adjacent to the area in question (spatially correlated or local variation). It is a fully Bayesian model which has been used substantially for disease mapping since it was introduced by Besag, York and Mollie in 1991. Under the BYM model, the smoothed SIR/SMR (or relative risk) is implemented using Gibbs sampling within WinBUGS. The sample values for the parameter of interest ( i) obtained by running this model in WinBUGS range in value within each LGA. This range of values form what is known as the posterior distribution of i, which represents the expected distribution of the SIR/SMR for each LGA when adjusted (smoothed) for the two types of variability mentioned above. For each LGA the mean of this posterior distribution was used as the best estimate of the smoothed SIR/SMR, and the proportion of the probability distribution above or below unity (one) was used to estimate the statistical significance of the small area estimate relative to the state average. It should to be noted that the posterior distribution is dependent upon the expected number of cases: the higher the expected number of cases the smaller will be the standard error of the distribution, and hence the posterior distribution of i will be 'tighter' around its mean.

A Bayesian 95% credible interval, which is obtained by selecting the 2.5th and 97.5th percentiles of the posterior distribution, is analogous to the more common 95% confidence interval used in frequentist-based analyses. It can be interpreted as the range in which 95% of the SIR/SMR estimates are located.

As noted previously, the proportion of the posterior probability distribution that lie above or below unity (one) was used to estimate the statistical significance of the small area estimate relative to the state average. A two-sided test was used, so that if the proportion of the posterior distribution above or below unity was less than 0.025, then that area was considered to have significantly decreased or increased risk at the 5% level of significance respectively.

All tabular output has been grouped by health area. As Sydney LGA is split between Sydney South West and South East Sydney and Illawarra AHS, results for this LGA are reported for both health areas. The columns in the output present the smoothed number of hospitalisations or deaths per year (suppressed if this was less than five); the smoothed SIR/SMR; the upper and lower 95% credible interval endpoints for the smoothed estimates of SSR/SMR; and the level of significance in relation to the state average indicated as follows:

3.7.1.2 Smoothing of estimates for binary-type indicators

These indicators do not use a population-based denominator. Usually the cases or events of interest are a subset of all cases, so the denominator is obtained from the same source as the cases of interest. Most of these indicators are binary in nature, for example smoking in pregnancy, where the mother either answers yes or no to the question 'did you smoke at all during pregnancy?', or else are made into a binary variable, for instance whether antenatal care was commenced before 20 weeks or not. Because of the binary nature of these indicators, the most appropriate model is the logistic model.

The smoothing used for these indicators is obtained by modelling the data using a binomial distribution with a logit link function (Lawson et al, 2003). The model still incorporates both spatially correlated (local variation) and uncorrelated variation (global variation), however results are not age-standardised. This model was also implemented using Gibbs sampling within WinBUGS. The sample values for the parameter of interest (smoothed proportion) range in value within each LGA. This range of values forms a distribution known as the posterior distribution of the smoothed proportion. For each LGA, the mean of the posterior distribution is used as the best estimate of the smoothed proportion.

In order to facilitate comparisons between areas and make mapping consistent with the population-based indicators, a posterior distribution for the prevalence ratio was calculated for each LGA by dividing sampled values of the smoothed proportion by the overall proportion for the state. The proportion of the posterior distribution of the smoothed prevalence ratio above unity was used as an estimate of the significance of the small area estimate relative to the state average. It should again be noted that the posterior distribution are dependent upon the total number of cases: the higher the total number of cases in an LGA the smaller the standard error of the distribution, and hence the posterior distribution of the smoothed prevalence ratio will be 'tighter' around its mean.

Ninety-five percent credible intervals for smoothed prevalence ratios and estimates of significance were obtained using methods similar to those discussed in 3.7.1 above. All tabular output included the smoothed number of cases per year (suppressed if this was less than five); the smoothed proportion; the smoothed estimate of the prevalence ratio; the lower and upper 95% credible interval endpoints for the smoothed prevalence ratio; and the level of significance in relation to the state average (using the same symbols as for 3.7.2).

3.8 Projected numbers and rates

Projected numbers and rates were estimated based on past trends. The base data used for hospitalisations was from 1st July 1998 onwards, because the introduction of ICD-10 coding of diagnoses and counts based on episode of care caused a break in trends for some indicators at that date. For all other variables, data for the last 20 years were used as the base data for predictions. In all cases, it was assumed that current trends will continue. No allowance was made for possible future changes in treatment regimens.

For each sex and age group combination, the number of future events (deaths, hospitalisations or cases)was estimated by fitting a generalised linear model with events as a linear function of year. A poisson distribution was assumed for the events. The model was fitted using a log link function and the log of the etimated residential population as an offset variable, using the Genmod procedure in SAS for Windows Version 9.1.3 (SAS, 2005). The projected counts for each age group were used to calculate projected age-standardised rates, and combined to calculate total counts for males, females and persons.

Population projections were obtained from the Transport and Population Data Centre, Department of Planning and are based on 2001 Census counts.

3.9 Record Linkage: the NSW Inpatient Statistics Collection or the NSW Admitted Patient Data Collection linked internally

Records within each year of the hospitalisation data were internally linked to group hospital records of patients and subsequently ascertain a number of persons hospitalised with a particular disease in each year.

The NSW Inpatient Statistics Collection's records were linked between 1996-97 and 1999-00. This record linking was partially automated using the Automatch system and was performed in the NSW Department of Health. The details of the method are explained below.

The hospitalisation data between July 2000 and the latest year available for linkage were linked by the Centre for Health Record Linkage (CHeReL). Details of the method are included below. The name NSW Inpatient Statistics Collection was changed to the NSW Admitted Patient Data Collection in 2005, but the name of the HOIST datasets has been preserved and the distinction between linked sets is made by referring to CHeReL: ISC2ISC contains years 1996-97 to 1999-00 and ISC2ISC (CHeReL) 2000-01 to the latest one.

3.9.1 Record linkage: ISC2ISC 1996-97 to 1999-00

Datasets for each financial year were linked separately. Record linking was partially automated using the Automatch system. Patient names were not available.

Prior to matching, addresses from the data were standardised using Autostan.

Match parameters were specified for eight linkage passes; records remaining unlinked from pass 1 were available for linkage in pass 2, and so on. For each pass, blocking variables were first defined to sort the data. Blocking variables included hospital code, medical record number, alternative hospital code, date of birth, year of birth, month of birth, day of birth, sex, locality, street name and postcode (the last three as soundexes).

Automatch sorts each input file by the blocking variables, and then compares records within each block, based on values of the matching variables. Matching variables comprised date of birth, country of birth, sex, insurance status, language spoken at home, marital status, Aboriginality, an array of admission date and separation date, postcode, hospital code, medical record number and standardised components of addresses.

The UNDUP program in Automatch was chosen for the linkage. Automatch selects a master record (also known as a match pair) that has the highest weight within a set of matched records. The master record is the most complete record, since missing values in a variable have less weight than variables with non-missing values. Duplicate records are records belonging to the master record. The master record and its set of duplicates are known as a pseudo match. Records that are not linked are called residuals and are flagged by 'AR' on the variable 'rlt_isc'.

After the automated record linkage, a match set number (mset or hsetp1) is generated for a set of matched records for the person and unlinked records. Records with match probability between the upper match threshold and lower threshold had to be manually reviewed, where a human decides if the pair of records belong to the same individual or not.

The linked data were also checked for false matches.

3.9.2 Record linkage: ISC2ISC from July 2000

The Centre for Health Record Linkage (CHeReL) has carried out internal linkage of the Admitted Patient Data Collection (APDC), formerly known as the Inpatient Statistics Collection (ISC), episodes of care datasets.

The CHeReL is a collaborative venture funded by eight member organisations (ACT Health, the NSW Clinical Excellence Commission, the Cancer Institute NSW, NSW Department of Health, the Sax Institute, the University of Newcastle, the University of New South Wwales and the University of Sydney) and hosted by the Cancer Institute NSW. The CHeReL provides a mechanism for access to linked health data for research. It uses probabilistic record linkage techniques to link personal information from a defined set of health-related datasets (including hospital inpatient and emergency department data, cancer registry data, birth and death data) to create a Master Linkage key, consisting of 'pointers' to records for persons on the source databases.

For specific projects, and subject to relevant ethics committee and data custodian approval, the CHeReL supplies lists of specified records to data custodians, who then provide approved data to the project investigators. Subject to relevant ethics committee and data custodian approval, the CHeReL also carries out ad hoc linkages of other health-related datasets.

To create ISC2ISC (CHeReL) datasets, first the APDC datasets were compiled into text files by financial year containing encrypted record ID (unique across years), name (patients' names have been available on the APDC from 1 July 2000), address, hospital code, transfer to and transfer from hospital codes, Medical Record Number, date of birth, date of admission, date of separation, sex and death date (derived from date of separation where death was reported as the mode of separation). These files were forwarded to the CHeReL. The details of hospitalisation are not made available to the CHeReL to conduct linkage.

Linkage was carried out at the CHeReL using ChoiceMaker software. ChoiceMaker uses 'blocking' and 'scoring' to identify definite and possible matches. During blocking ChoiceMaker searches the target datasets for records that are possible matches to each other. There are two types of blocking. The exact blocking algorithm requires records to have the same set of valid fields and the same values for these fields. The automated blocking algorithm builds a set of conditions that are used to find as many as possible records that potentially match each other. Scoring employs a combination of a probabilistic decision, which is computed using a machine learning technique, and absolute rules, which include upper and lower probability cut-offs, to determine the final decision as to whether each potential match denotes or possibly denotes the same person. Upper and lower probability cut-offs were determined iteratively until acceptable accuracy rates were obtained. Human review was performed on groups with match probability between 0.236 and 0.73. The accuracy was: false positive rate 3/1,000; false negative rate less than 1/1,000.

The APDC data form part of the CHeReL Master Linkage Key. Each record in the Master Linkage Key is assigned a record number and a CHeReL Person ID to allow linked records for the same individual to be identified and extracted.

The CheReL provided project keys, comprising the original encrypted record ID and a Project Person Number, to the Centre for Epidemiology and Research. The record IDs were decrypted and the Project Person Numbers were attached to the full datasets. The Project Person Number is used to extract records referring to the same person.

The linked data do not contain records for interstate hospitalisations. Interstate data do not have personal identifiers such as date of birth, medical record number, name and address, and are therefore unable to be linked.

The linked data were split by financial year of separation.

In this report the data linked within individual years 1996-97 to the latest available were used in some indicators and data linked across years covered by ISC2ISC (CHeReL) in others, which is then acknowledged in the accompanying notes.

4. Methods used for specific chapters and topics

4.1 Health-related behaviours and other chapters - Deaths and hospitalisations attributable to health risks

Estimates of the numbers and rates of deaths and hospitalisations attributable to the use of tobacco, alcohol, to high body mass and other risk factors used age and sex-specific aetiologic fractions developed by the School of Population Health, University of Queensland and the Australian Institute of Health and Welfare and published in 2007 (Begg et al., 2007).

The contribution of 14 health risks to the total burden of disease was assessed by the School of Population Health, using methods developed by the WHO Comparative Risk Assessment project (WHO, 2004). Earlier work by English and colleagues (English et al., 1995) was also used with reference to risks from the use of drugs and alcohol by the researchers from the School of Population Health. The main elements of the methodology are the prevalence of exposure to a health risk in a population and information on the risk of disease, injury or death from this exposure, which is derived from meta-analysis of published scientific literature. Calculations result in estimates of the proportions of cases of specific diseases and injuries that could be attributed to each risk factor.

For this report, electronic files of the aetiologic fractions developed by Begg and colleagues was obtained directly from the School of Population Health, University of Queensland. The disease and injury groupings used in this file were defined using coding developed for the Burden of disease study. There are teo steps in applying the aetiologic fractions:

(a) ill-defined categories (e.g. heart failure, unspecified diabetes mellitus and injuries with unspecified intent) are redistributed into specific categories based on other information in the record and/or on a pro rata basis;

(b) the aetiologic fractions are applied to categorised records.

4.2 Rural and remote populations

The chapter on rural and remote populations and some indicators in other chapters present a range of health issues for NSW according to ARIA+, an enhanced Accessibility-Remoteness Index of Australia classification.

In ARIA+ the remoteness index value is based on road distance to each of five categories of 'service centre'. The service centre categories are based on population size, with the smallest centres having populations of 1,000-4,999. Localities with populations of greater than 1000 persons are considered to contain at least some basic level of services (e.g. health, education, or retail) (GISCA, 2001). Service centres with larger populations are assumed to contain a greater level of service provision. ARIA+ scores range from 0 to 15, and are grouped into remoteness categories. There are five classes of remoteness: major cities, inner regional, outer regional, remote and very remote (AIHW, 2004). Census Collection Districts (CDs) are assigned Australian Standard Geographical Classification (ASGC) remoteness categories based on the average ARIA+ score within the CD. Statistical Local Areas (SLAs) are then classified by the proportion of the population living in CDs in each ASGC remoteness category.

4.3 Country of birth

In September 2008, the six countries that comprise the Former Yugoslavia were: Bosnia and Herzegovina, Croatia, the Former Yugoslav Republic of Macedonia, Montenegro, Serbia and Slovenia. An additional category Former Yugoslavia, not further defined (nfd), is also included in analyses by country of birth. Depending on the data source and the span of years, the Former Yugoslavia, not further defined (nfd) category may comprise all data concerning all six countries or only a small proportion of data that cannot be defined more precisely. Footnotes describe specific cases. Persons born in Yugoslavia comprised 1% of population of NSW and Australia at the time of the breakup of Yugoslavia in the early 1990s (ABS, 2008).

4.4 Socioeconomic status

Methods used in the chapter on socioeconomic status are described below.

4.4.1 Socioeconomic status measures

The measure of socioeconomic status (SES) used in this report was the Index of Relative Socio-Economic Disadvantage (IRSD), which is one of four Socio-Economic Indexes for Areas (SEIFA) currently produced by the Australian Bureau of Statistics (ABS) using census data (ABS 1998; ABS 2003; ABS 2008). Each census, the ABS assigns an IRSD score to non-overlapping geographical areas covering all of Australia calculated from various socioeconomic characteristics of the people living in areas. In the IRSD, these characteristics relate to occupation, education, non-English speaking background, indigenous origin, and the economic resources of the household.

The ABS has released IRSD scores after the last five censuses. The methods used to calculate IRSD scores were similar in 1986, 1991 and 1996, but changed in 2001 and 2006. The major change in 2006 was that the census data used in the calculation of the IRSD was based on people's usual area of residence rather than their location on census night (place of enumeration). IRSD scores are an ordinal measure with a mean of 1000 and standard deviation of 100 for Australia, based on the index scores of all Census Collection Districts (CDs) in Australia. Scores for larger geographic areas such as Statistical Local Areas (SLAs) and Local Government Areas (LGAs) are population-weighted averages of scores in constituent CDs. The areas can be ranked by IRSD score and compared by rank between censuses, but other arithmetic comparisons using scores are not valid. For example, the IRSD score for NSW from the 1996 census was 1006, which means that the SES of NSW was slightly better than Australia as a whole. The score for NSW in 1991 was 1002; however, that does not mean that NSW was better off in 1996 than in 1991 as the scores were calculated based on socioeconomically different Australian populations.

In this report, the NSW population was divided into five groups based on the IRSD scores of their SLA of residence. This means that SLAs were sorted by IRSD score and assigned to population-weighted quintiles, each containing close to one-fifth of the total population. For all statistical analyses in the SES chapter the quintiles were divided into three groups: the lowest SES population-weighted quintile, the highest SES population-weighted quintile, and the rest of the population, comprising the remaining three population-weighted quintiles. In tables however, numbers and rates are reported separately for each population-weighted quintile.

4.4.2 Poisson regression models

Poisson regression models (Armitage et al., 2002) were used to investigate differences in trends in mortality, hospitalisation and teenage pregnancy rates over time between SES groups. For each indicator, age-adjusted rates were modelled by sex to ascertain the magnitude and statistical significance of changes in relative health gaps between SES groups over time.

Models were fit with counts of outcomes as response variables, offset by the log of the relevant NSW population. All models included terms for age (18 categories from 0-4 to 85+ years), SES group, year, and an interaction between SES group and year. To adjust for over-dispersion, model standard errors were scaled using the Pearson chi-square dispersion parameter. Relative rates of change in outcomes over time were calculated for each SES group by taking the exponent of appropriate combinations of year and SES*year parameter estimates. The statistical significance of differences in relative rates of change over time was assessed by testing for differences between the fitted slopes of each SES group. All models with fit using the GENMOD procedure in SAS 9.1.3 (SAS, 2005).

4.4.3 Linear regression models

Linear regression models (Neter et al., 1996) were used to investigate differences in trends in life expectancy and prevalence estimates over time between SES groups. All outcomes were log-transformed and modelled via ordinary least squares regression with terms for SES group, year and an interaction between SES group and year. A similar process to that described for Poisson regression models in 4.4.2 was used to estimate relative rates of change over time and assess the statistical significance of differences between the fitted slopes of each SES group. All models were fit using the REG procedure in SAS 9.1.3 (SAS, 2005).

4.5 Potentially avoidable mortality

The method used to calculate avoidable mortality in NSW in this report is based on a method described in ANZ Atlas of avoidable mortality (Page et al., 2006), which in turn is a revision of the original set of conditions and methodology developed by Tobias and Jackson (Tobias and Jackson, 2001). The Atlas is an authoritative source of information on avoidable conditions for Australia and New Zealand.

Avoidable deaths are those attributed to conditions that are considered preventable or otherwise avoidable through earlier intervention or action and which occur before age 75 years. Following the definition, avoidable deaths are further sub-categorised into preventable and from causes amenable to health care. To simplify and make the categorisation more stable over time, each condition is ascribed totally to the preventable or amenable group, depending on which type of intervention plays greater role in making the condition 'avoidable'. Only three conditions - diabetes, ischaemic heart disease and cerebrovascular diseases - have been placed equally apportioned to both groups.

The codes used to define avoidable mortality groups, along with the sub-categorisation can be found in the disease and procedure codes section of this report.

4.6 Ambulatory care sensitive conditions

The method used to calculate avoidable hospitalisations used the concept of ambulatory care-sensitive (ACS) conditions. These are hospitalisations that could have been avoided through the use of preventive healthcare or early disease management given in an ambulatory setting, such as by a general practitioner or community health centre.

The categories used for the ambulatory care-sensitive conditions are based on those used by the Victorian Government Department of Human Services (VGDHS, 2004), which have been reviewed by the Public Health Information Development Unit (Page, 2007).

The information presented in this report differs from information presented in earlier editions. In 2006 in NSW the coding of diabetes was changed to include diabetes in primary diagnosis only, which resulted in fewer cases of diabetes and therefore chronic conditions overall. In 2007, urinary tract infections were included, which increased the number of cases in acute conditions category, and the coding of cellulitis was aligned with national standards.

The codes used can be found in the disease and procedure codes section of this report.

4.7 Diabetes-related deaths

The term 'diabetes-related death' is used in this report to refer to deaths where diabetes was recorded as the underlying cause of death, or where diabetes was recorded as an associated cause of death and the underlying cause of death was one of commonly recognised diabetes complications. These complications were myocardial infarction, ischaemic heart disease*, stroke or sequelae of stroke*, heart failure*, sudden death (cardiac arrest), peripheral vascular disease, kidney disease, hyperglycaemia and hypoglycaemia (Dixon at al., 2005).

The reason for this approach is that, more than other disorders, diabetes often causes death indirectly because it is a strong risk factor for common causes of death such as heart and kidney disease, and stroke. These complications are likely to appear as the underlying cause of death, the basis for official mortality statistics. If only cases where diabetes as the underlying cause were counted, it would lead to considerable underestimates of diabetes' contribution to death in Australia (Dixon et al., 2005).

The concept of 'diabetes-related deaths' is based on the definition of 'death related to diabetes' used in the United Kingdom Prospective Diabetes Study (UKPDS) since 1998. The UKPDS definition has been modified by diabetes specialists on the National Diabetes Data Working Group, associated with the Australian Institute of Health and Welfare (AIHW), to include additional conditions (ischaemic heart disease, stroke or sequelae of stroke, heart failure, marked above with asterisks) (Dixon at al, 2005).

For the full list of codes of included conditions see the disease and procedure codes section of this appendix.

4.8 Psychological distress

The K10 (Kessler and Mroczek, 1992, 1994; Kessler et al., unpublished) was included in the 1997, 1998, 2002 and 2005 NSW Population Health Surveys as a relatively short measure of psychological distress that allowed comparison against international survey data and validation against concurrent diagnostic data in the National Survey of Mental Health and Wellbeing (NSMHW) (ABS, 1997; Andrews and Slade, 2001). The K10 (and a briefer K6 measure) were specifically designed for use in the 'core' of the annual US National Health Interview Survey (N=50,000 aged 15+) when it was redesigned for use from 1997 onwards. The K6 has also been used in the biennial Canadian National Population Health Survey (panel survey, N>17,000 aged 12+; 1994-95; 1996-97; 1998-99; 2000-01) and has been replaced by the K10 in the Canadian Community Health Survey from September 2000 (Statistics Canada, 2002).

The K10 is currently being used in a series of surveys similar to the Australian NSMHW in 20 countries, under the auspices of the World Health Organization (WHO). These surveys have a total sample size of about 200,000. The WHO regions surveyed include North America (Canada and the United States), Latin America (Brazil, Colombia, Mexico and Peru), Europe (Belgium, France, Germany, Italy, The Netherlands, Spain, and The Ukraine), the Middle East (Israel), Africa (South Africa), and Asia (China, India, Indonesia, Japan, and New Zealand) (Kessler et al., 2000).

The K10 measure is a 10-item self-report questionnaire intended to yield a global measure of 'psychological distress' based on questions about the level of restlessness, anxiety, and depressive symptoms in the most recent four-week period. It is designed to span the range from few or minimal symptoms through to extreme levels of distress, which is an essential feature of an instrument for use in population studies. Thus the K10 contains low-threshold items, that many people may endorse, through to high-threshold items that very few will endorse. Overall, the item-response scale is designed to yield most precision around the 90th to 99th percentile of the general population.

For each item there is a five-level response scale based on the amount of time (from none through to all) during a four-week period when the person experienced the particular problem. In NSW use, there are also four follow-up questions, that aim to quantify the level of disability resulting from the feelings of distress, the health service usage resulting from the distress, and the extent to which the distress is believed to be mainly due to physical health problems.

Scoring of the raw questionnaire assigns between 1 and 5 points to each symptom in the direction of increasing problem frequency. Thus the raw score range is from 10 (all responses to all questions are 'none of the time') through to 50 (all responses to all questions are 'all of the time'). Low scores indicate low levels of psychological distress and high scores indicate high levels of psychological distress (ABS, 2003).

The creators of the K10 have not yet published details on scoring the scale and there has been no international standard for determining cut-off points for low, medium and high levels of psychological distress (ABS, 2003). Various interpretations of scoring have been used in Australia and worldwide. Recently, and following the advice of the K10 originators, NSW adopted a four-level approach to illustrate prevalence and severity. The four levels are given in Table 2.

These cut-off scores were previously used in the 2000 Health and Wellbeing Survey (conducted in Western Australia) and the ABS 2001 National Health Survey Summary of Results Publication (ABS, 2003). The adoption of the above scores in NSW ensures comparability of the NSW results with national and, increasingly, international data.

5. Area Health Service boundaries

In July 2004, the Minister for Health announced a new area health structure in NSW. The restructure is largely a merger of the 17 Area Health Services into 8 new Area Health Services. The new administrative structure was in place by 1 January 2005. A list of the mergers is given in Table 3. Some localities within the Statistical Local Areas of Greater Taree, Greater Lakes, Gloucester and Lithgow moved to the new Area Health Services independently of the main mergers.

In this report, the 2005 Area Health Services boundaries have been used. A list of local government areas by 2005 Health Area is also provided in this appendix.

6. Quality assurance process

The preparation of this report involved complex data processing and manipulation. The following steps were taken to minimise errors.

• most analyses used a single, shared suite of datasets, contained on the HOIST system. The datasets on HOIST are checked against the original source data to ensure their fidelity. Sources for all data used are described in the footnotes.
• all graphs and tables were produced using SAS programs that can be audited, rather than using interactive data manipulation facilities such as spreadsheets that are much more difficult to check.
• the SAS programs directly created the Web pages for the online version of the report as well as tables and graphs. This minimised the possibility of transcription and typographical errors.
• every SAS program used in the production of this report was checked by someone other than the person who originally wrote it. Items such as the correct specification of ICD codes and correct selection of numerator and denominator data were systematically checked as part of this audit process.
• complex parts of the SAS programs were abstracted as a common, shared set of SAS 'macros' (callable subroutines). These macros, which were employed for operations such as imputation, direct standardisation and production of custom graph formats, were subject to rigorous testing before they were used.
• all results were checked against other, comparable sources, wherever possible.

For more information

Andrews G, Slade T. Interpreting scores on the Kessler psychological distress scale (K10). Aust N Z J Public Health 2001, 25:494-497.

Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research, 4th edition. Oxford: Blackwell Science, 2002.

Australian Bureau of Statistics and Australian Institute of Health and Welfare. The health and welfare of Australia's Aboriginal and Torres Strait Islander peoples, 2008. ABS Cat. No. 4704.0. AIHW Cat. No. IHW 21. Canberra: ABS, 2008. Available at www.abs.gov.au/AUSSTATS/abs@.nsf/mf/4704.0

Australian Bureau of Statistics.Deaths, Australia 2006. Cat. No. 3302.0. Canberra: ABS, 2008. Available at www.abs.gov.au/ausstats/abs@.nsf/mf/3302.0

Australian Bureau of Statistics. Information paper: 1996 census socioeconomic indices for areas. Cat. No. 29120. Canberra: ABS, 1998.

Australian Bureau of Statistics. Information paper: 2001 Socioeconomic indexes for areas. Cat. No. 2039.0. Canberra: ABS, 2003.

Australian Bureau of Statistics. Information paper: 2006 Socioeconomic indexes for areas. Cat. No. 2039.0. Canberra: ABS, 2008.

Australian Bureau of Statistics. Mental health and wellbeing of adults, Australia 1997. Cat. No. 4360.0. Canberra: ABS, 1997.

Australian Institute of Health and Welfare. Rural, Regional and Remote Health: A guide to remoteness classifications AIHW Cat. PHE 53. Canberra: AIHW, 2004. Available at www.aihw.gov.au

Australian Institute of Health and Welfare Dental Statistics and Research Unit. Child Dental Health Survey, New South Wales, 1999. Adelaide: DSRU, 2001.

Begg S, Vos T, Barker B. et al. The burden of disease and injury in Australia, 2003. AIHW Cat. No. PHE 82. Canberra: AIHW, 2007. Available at www.aihw.gov.au/publications/index.cfm/title/10317

Besag J, York J, Mollie A. Bayesian image restoration with two applications in spatial statistics. Annals of the Institute of Statistical Mathematics 43, 1-59, as quoted in Lawson et al (2003).

Booth M, Okely AD, Denney-Wilson E, Hardy L, Yang B, Dobbins T. NSW School Physical Activity and Nutrition Survey (SPANS) 2004: Full Report. Sydney: NSW Department of Health, 2006. Available at www.health.nsw.gov.au/pubs/2006/spans_full.html

Booth M, Okely T, Denney-Wilson E, Hardy L, Yang B, and Dobbins T. NSW School Physical Activity and Nutrition Survey (SPANS) 2004: Summary Report. Sydney: NSW Department of Health, 2006. Available at www.health.nsw.gov.au/pubs/2006/spans_summary.html.

Centre for Health Record Linkage available at www.cherel.org.au

Chiang CL. The life table and its applications. Malabar: Krieger, 1984.

Commonwealth of Australia. 2007 Australian National Children's Nutrition and Physical Activity Survey - Main Findings. Canberra: Commonwealth of Australia, 2008. Available at www.health.gov.au/internet/main/publishing.nsf/Content/health-pubhlth-strateg-food-monitoring.htm#07survey

Dixon T, Webbie K. Diabetes-related deaths 2001-2003. Bulletin No. 32. AIHW Cat. No. AUS 69. Canberra: AIHW, 2005. Available at www.aihw.gov.au/publications/index.cfm/title/10221

Dobson A, Kuulasmaa K, Eberle E, Schere J. Confidence intervals for weighted sums of Poisson parameters. Statistics in Medicine 1991; 10: 457-462.

English DR, Holman CDJ, Milne MG et al. The quantification of drug-caused morbidity and mortality in Australia. Canberra: Commonwealth Department of Human Services and Health, 1995.

Ezzati M, Lopez AD, Rodgers A, Murray CJL (eds). Comparative quantification of health risks: Global and regional burden of disease attributable to selected major risk factors. Geneva: World Health Organization, 2004.

Graham-Clarke P, Howell S, Bauman A, Nathan S. NSW Health Promotion Survey 1994 Technical Report. Sydney: National Centre for Health Promotion and NSW Department of Health, 1995.

Holman CDJ, Armstrong BK, Arias LN et al. The quantification of drug-caused morbidity and mortality in Australia, 1988. Canberra: Commonwealth Department of Human Services and Health, 1990.

Kessler RC, Andrews G, Colpe LJ, Hiripi E, Mroczek DK, Normand S, Walters EE. Short screening scales to monitor population prevalences and trends in nonspecific psychological distress. (unpublished).

Kessler RC, Costello EJ, Merikangas KR, Ustun TB. Psychiatric Epidemiology: Recent Advances and Future Directions. Chapter 5 in Manderscheid RW, Henderson MJ (editors) Mental Health, United States, 2000. Rockville, Maryland: US Department of Health and Human Services, 2000.

Kessler R, Mroczek D. An update of the development of mental health screening scales for the US National Health Interview Survey. Ann Arbor MI: Survey Research Centre of the Institute for Social Research, University of Michigan, Memo dated December 22, 1992.

Kessler R, Mroczek D. Final versions of our Non-Specific Psychological Distress Scale. Ann Arbor MI: Survey Research Centre of the Institute for Social Research, University of Michigan, Memo dated March 10, 1994.

Last JM. A Dictionary of Epidemiology, 4th edition. New York: Oxford University Press, 2001.

Lawson AB, Browne WJ, Rodeiro, CL. Disease Mapping with WinBUGS and MLwiN. Chichester: John Wiley & Sons, 2003.

Muscatello D, Travis S. Using the International Classification of Diseases with HOIST. NSW Public Health Bulletin 2001; 12 (11): 289-293.

National Centre for Classification in Health. ICD-10-AM, 2nd edition. Sydney: National Centre for Classification in Health, 2000.

National Centre for Social Application of Geographic Information Systems (GISCA). About ARIA+ (Accessibility/Remoteness Index of Australia). Adelaide: GISCA, 2001. Available at www.gisca.adelaide.edu.au/products_services/ariav2_about.html..

National Coding Centre. The Australian version of the International Classification of Diseases, 9th revision, clinical modification (ICD-9-CM). Sydney: University of Sydney, 1996.

Neter J, Kutner MH, Machtsheim CJ, Wasserman W. Applied Linear Statistical Models, 4th edition. McGraw Hill, 1996.

NSW Centre for Overweight and Obesity. NSW School Physical Activity and Nutrition Survey (SPANS) 2004: Short Report. Sydney: NSW Department of Health, 2006. Available at www.healthpromotion.com.au/Documents/CIM/spans_short_report.pdf

NSW Department of Health. The health behaviours of secondary school students in New South Wales 2002. NSW Public Health Bulletin 2004; 15(S-2).

NSW Health Survey Program website at www.health.nsw.gov.au/public-health/survey/hsurvey.html

Page A, Tobias M, Glover J, Wright C, Hetzel D, Fisher E. Australian and New Zealand atlas of avoidable mortality. Adelaide: PHIDU, University of Adelaide, 2006. Available at www.publichealth.gov.au/publications/atlas-of-avoidable-mortality-new-zealand.html

Page A, Ambrose S, Glover J, Hetzel D. Atlas of avoidable hospitalisations in Australia: ambulatory care-sensitive condition. Adelaide: PHIDU, University of Adelaide and AIHW, 2007. Available at www.publichealth.gov.au/publications/atlas-of-avoidable-hospitalisations-in-australia%3a-ambulatory-care-sensitive-conditions.htm

Population Health Division. New South Wales Older People's Health Survey 1999. NSW Public Health Bulletin 2000; 11 (5-7).

Population Health Division. Review of the Save Our Kids Smiles (SOKS) program Volume I: Report and Volume II: Technical Report. Sydney: NSW Department of Health, 2001.

Ridolfo B, Stevenson C. The quantification of drug-caused morbidity and mortality in Australia, 1998. AIHW Cat. No. PHE 29. Canberra: AIHW, 2001.

SAS Institute. The SAS System for Windows version 9.1.3. Cary, NC: SAS Institute, 2005.

Tobias M, Jackson G. Avoidable mortality in New Zealand, 1981-97. Aust N Z J Public Health 2001; 25: 12-20.

Victorian Government Department of Human Services. The Victorian Ambulatory Care Sensitive Conditions Study, 2001-02. Melbourne: VGDHS, 2004. Available at www.dhs.vic.gov.au/health/healthstatus/acsc/finalreport.htm.

Williamson M, Baker D, Jorm L. The NSW Health Survey Program: Overview and methods 1996-2000. NSW Public Health Bulletin 2001; 12 (5-x).

WinBUGS Version 1.4.3 Imperial College and MRC, UK, WinBUGS 2007.

World Health Organization. International Classification of Diseases, 9th revision. Geneva: WHO, 1977.

World Health Organization. International Statistical Classification of Diseases and Related Health Problems, 10th revision. Geneva: WHO, 1992.

Print version with data

Although this page can be printed directly from your Web browser, a higher quality version of this entire page (graph, table and text) is available as an Acrobat PDF file which can be printed or viewed on screen using free software.

Copyright notice

This work is copyright NSW Department of Health, 2006. It may be reproduced in whole or in part, subject to the inclusion of an acknowledgement of the source. Commercial usage or sale is prohibited.

Suggested citation

Population Health Division. The health of the people of New South Wales - Report of the Chief Health Officer. Sydney: NSW Department of Health. Available at: www.health.nsw.gov.au/publichealth/chorep/. Accessed (insert date of access).

Produced by

Centre for Epidemiology and Research, Population Health Division, NSW Department of Health.

Last updated on 16 January 2009