 |
This methodology section provides a brief overview of key issues methods, and sources. A detailed methodology report is available.51 For further information, contact Population_Epidemiology@health.qld.gov.au.
Demographic data were derived from ABS and are used under the Creative Commons Attributions 2.5 Australia licence. Copyright in ABS data vests with the Commonwealth of Australia, www.abs.gov.au
Death data were derived from the Queensland Registry of Births, Deaths and Marriages, where cause of death and usual residential address were recorded on death certificates. Unless otherwise specified, all disease specific death data in this report were derived using the underlying cause of death and refer to year of registration of death of Queensland residents in the relevant year. The 2007 ABS death data release for this report is considered preliminary, as at the time of release the improved processing of coroner certified deaths had not been conducted, and is therefore subject to change.59 However for reporting suicide in Chapter 4, the revised 2007 death data released in 2008 (that is, coroner certified revisions) were included.52 In addition, due to limitations at the time of extraction, for 2007 the registered deaths of Queensland residents in New South Wales, Victoria and Tasmania are not included (a small number of exceptions are noted in Table 1). Based on a later release from ABS this undercount was 369 deaths or 1.4% of total.52 Data were coded using the ICD version 10 (ICD-10), with code sets for specific conditions consistent with Queensland Health guidelines and aligned with national reporting.493 At the time of extraction, 2007 jurisdictional death data was not available. International comparisons may be limited by inconsistencies in coding between countries.
Hospitalisation data (separations or episodes of care) were derived from the Queensland Hospital Admitted Patient Data Collection, including admissions of Queensland residents to private and public hospitals, but excluding admissions to public psychiatric hospitals, admissions for unflagged newborns, organ donors or boarders, and where patient sex is indeterminate/ intersex. All disease-specific hospital separations were derived using the principal diagnosis of inpatient episodes of care unless otherwise specified. The external causes of injury and poisoning are reported. All separations were coded using the ICD version 10 Australian Modification (ICD-10-AM) using standard code sets.493
Rates for all diseases and conditions are reported as age standardised rates. Adjustment of rates for age allows comparisons between populations that have different age structures, for example between youthful and older communities. The direct standardisation method was used in this report, where age specific rates are multiplied against a standard reference population. The 2001 Australian population (persons) was used as the standard population. For all OECD comparisons, the OECD population of 1980 was used as the reference population. The 95% CIs for rates were estimated by assuming Normal approximations to the Poisson distribution.
The projected admission rates are based on linear regression of the most recent seven years of statewide admission.494 For each Enhanced Service Related Groups/Age Group/Stay Type combination, the projected population of a residence is multiplied by the projected state admission rate to yield an expected number of admissions.118
Survey data are reported as percentage and 95% CIs. Unless otherwise noted, all survey data refer to self report. Sources external to Queensland Health are cited. More information about Queensland Health collections is available.51
Cancer incidence data were derived from the Queensland Cancer Registry unless otherwise cited.
Communicable disease notifications were derived from the Queensland Health Notifiable Conditions System and vaccination data from the Australian Childhood Immunisation Register.
Perinatal and maternal data were derived from Queensland’s Perinatal Data Collection.
Cancer screening data were derived from screening databases (Pap Smear Register, BreastScreen Queensland and bowel screening database).
Socioeconomic disadvantage/advantage was determined using the ABS Index of Relative Socioeconomic Advantage/Disadvantage and related Indices (SEIFA).495 Quintiles were generated by Queensland Health using population weighted Statistical Local Areas, SEIFA scores. Accessibility/remoteness was determined using the Accessibility/ Remoteness Index of Australia (ARIA).496 The Remoteness Area index was used: Major cities, Inner regional, Outer regional, Remote, Very remote. Excess deaths due to socioeconomic disadvantage or remoteness were calculated using a modified version of the methodology described by the AIHW.51, 472
