National Integrated Health Service Information (NIHSI) version 2.0
Data Quality Statement Attributes
Identifying and definitional attributes | |
Metadata item type: | Data Quality Statement |
---|---|
METEOR identifier: | 789452 |
Registration status: | AIHW Data Quality Statements, Standard 21/03/2024 |
Data quality | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Data quality statement summary: | Summary
Description The NIHSI v 2.0 is an enduring linked data asset that brings together the following data sets:
Data on Indigenous status are mainly sourced from the national hospitals data collections. The inclusion of the Voluntary Indigenous Identifier has not yet been agreed. The NIHSI v 2.0 can be accessed through the AIHW Research Only Network (RON) and the Department of Health and Aged Care's Enterprise Data Warehouse (EDW). | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Institutional environment: | The Australian Institute of Health and Welfare (AIHW) is an independent corporate Commonwealth entity under the Australian Institute of Health and Welfare Act 1987 (AIHW Act), governed by a management board and accountable to the Australian Parliament through the Health portfolio. The AIHW is a nationally recognised information management agency. Its purpose is to create authoritative and accessible information and statistics that inform decisions and improve the health and welfare of all Australians. Compliance with the confidentiality requirements in the AIHW Act, the Privacy Principles in the Privacy Act 1988 (Cth) and AIHW's data governance arrangements ensures that the AIHW is well positioned to release information for public benefit while protecting the identity of individuals and organisations. For further information see the AIHW website, which includes details about the AIHW’s governance and role and strategic goals. The NIHSI was approved by the Australian Health Ministers Advisory Council and is owned by the AIHW, the Commonwealth Department of Health and Aged Care, and state and territory health authorities. It is managed under the custodianship of the AIHW, in consultation with representatives from state/territory health departments, and the Commonwealth Department of Health and Aged Care. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Timeliness: | The NIHSI v 2.0 data set was released in December 2022 and made available to approved analysts. The NIHSI v 2.0 holds data from 1 July 2010 to 30 June 2020. Therefore, the timing of the data asset release was more than two years after the end of the reference period. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Accessibility: | NIHSI v 2.0 is available for access by Commonwealth and state/territory health departments, health portfolio agencies and the AIHW. The NIHSI v 2.0 can be accessed by named analysts on approved projects from these organisations. Projects proposed from the above organisations are assessed to be consistent with the AIHW Ethics Committee approval and the governance protocols by the AIHW’s Ethics Committee Delegate, and the AIHW Data Custodian to access NIHSI. Project proposals are then sent to the NIHSI Advisory Committee for approval. Project proposals will name the analysts undertaking the project. Only named analysts for approved projects can access the NIHSI. Projects using the NIHSI v 2.0 have access to all data from participating jurisdictions, except for projects that propose to use Queensland hospitals data and can only access that data upon receipt of explicit approval from the Queensland data custodian(s). The NIHSI can be accessed through secure access environments, namely the AIHW Research Only Network (RON) and in the Department of Health and Aged Care's Enterprise Data Warehouse (EDW). Access to these host environments is managed by the AIHW Data Custodian and in accordance with the governance protocols for NIHSI. Several published reports are available that have drawn upon data held in the NIHSI. All approved aggregate data released from the secure access environments satisfy output requirements and approval processes. All reports, publications and products are published and accessible from the AIHW website (www.aihw.gov.au/reports-data/nihsi/current-approved-projects). | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interpretability: | Information on the linked data sources for NIHSI v 2.0 is available on the AIHW website. Metadata information for each source collection are published in the AIHW's online metadata registry – METEOR, and the National Health Data Dictionary. METEOR and the National health data dictionary can be accessed on the AIHW website:
The National Death Index (NDI) is maintained by AIHW in the National Mortality Database (NMD). The data quality statements underpinning the AIHW NMD can be found in the Australian Bureau of Statistics (ABS publications): Further information can be found on the AIHW website on the National Mortality Database and National Death Index. Public hospital admitted patient episode data in the NIHSI are drawn from the National Hospital Morbidity Database (NHMD). The 2019–20 Admitted Patient Care National Minimum Data Set (NMDS) data items can be found under METEOR ID 699728. Please refer to METEOR for other years. Non-admitted patient emergency department care are patients registered for care in emergency departments. For further information, see Non-admitted patient emergency care NMDS 2019–20 in METEOR. Information on Medicare Benefits Schedule (MBS) data collection and Pharmaceutical Benefits Scheme (PBS) can be found on the AIHW website. Information on the National Aged Care Data Clearinghouse is available from National Aged Care Data Clearinghouse - Australian Institute of Health and Welfare (aihw.gov.au) | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Relevance: | The NIHSI is a linked person-level national information asset about the services provided by Australia's hospitals, medical and other services through the MBS, PBS, including RPBS, and residential aged care, linked to the National Death Index. The NIHSI v 2.0 holds data between 1 July 2010 and 30 June 2020, and can be used to produce outputs to inform medical research and health policy development, and the planning, monitoring and evaluation of health and aged care service delivery. NIHSI v 2.0 does not include Western Australia and Northern Territory hospitals data. The AIHW is continuing to negotiate with Western Australia and the Northern Territory to enable their hospitals data to be included in the NIHSI in the future. Coverage of admitted patient private hospitals and non-admitted patient care data within NIHSI v 2.0 is limited, subject on the availability of identifiers provided by states and territories to the AIHW for its data linkage activities to create the NIHSI. Only Victoria, Queensland and the Australian Capital Territory have private hospitals data included in NIHSI v 2.0. In addition, only Tasmania, Australian Capital Territory, Victoria, Queensland, and New South Wales have non-admitted patient care data in NIHSI v 2.0 (see table in the Quality Statement Summary for details of which years are included). As such, the scope of admitted patient private hospitals and non-admitted patient care records is underrepresented and varies across participating states and territories. Data on Indigenous status are sourced from the national hospitals data collections to create a derived Indigenous flag that represents an ever-identified Indigenous status. Only records for persons with a linked hospital record will have a derived Indigenous flag reported. Data on a person's usual residence varies across data collections included in the NIHSI, with the minimum geography available at the postcode level and either Statistical Area Level 2 (based on the ABS Australian Statistical Geography Standard (ASGS) Edition 2) or Statistical Local Area (based on the ABS Australian Standard Geography Classification) for earlier years of data. Geographical information about the hospital is reported based on state of service and not the state of the patient’s usual residence. Therefore, it may be difficult to identify accurate cross-border service provision when analysing patient pathways with the levels of geography information available in NIHSI. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Accuracy: | Data included in the NIHSI v 2.0 are sourced from administrative data collections designed for various purposes. As such, the AIHW has limited capacity to validate data quality. The data collection and cleaning processes varies across these collections, and the quality of the NIHSI v 2.0 will be subject to the quality of the data held in these source collections. To assist in interpreting hospitals data, more detailed information about the data sources, data quality, analysis methods, and the definitions and classifications used are available from About the data - Australian Institute of Health and Welfare. A data quality statement for the NDI is available from National Death Index (NDI), Data Quality Statement. Data quality information for the AIHW National Aged Care Data Clearinghouse and Aged Care Funding Instrument can be found in METEOR. Data on MBS and PBS/RPBS that may be useful to users can be found in METEOR. Linkage quality The quality of linkage depends on the coverage of identifiers available for each collection, and consistency with information held in the integrating spine (i.e., Medicare Consumer Directory (MCD) and National Death Index (NDI) were first linked to create the NIHSI linkage spine). This means that NIHSI v 2.0 is based on data that is linked to MCD and NDI based information. People who are not eligible for Medicare are not included in the spine. Some admitted patients may not be enrolled in or are not eligible for Medicare but will still be included in the National Hospital Morbidity Database (NHMD) such as international students or some overseas visitors who were admitted to public hospitals. For example, overseas visitors from New Zealand, Ireland, the United Kingdom, the Netherlands, Sweden, Finland, Norway, Italy, Malta, Belgium and Slovenia may receive public hospital care because Australia has Reciprocal Health Care Agreements with these countries. Over coverage in these cases may occur due to a lack of information or when the individuals leave Australia and are no longer considered as usual residents. This may mean that individuals may continue to be counted in the analysis after these individuals are no longer residents in Australia unless methods are applied to adjust for this over coverage. As such, under coverage or over coverage of different groups within the Australian resident population need to be considered in the analysis and interpretation of NIHSI data. As the NIHSI v 2.0 spine is made up of MCD and NDI records, we expect 100% linkage rates for MBS, PBS, and NDI records in NIHSI v 2.0. The table below shows percentage of linked and unlinked records in NIHSI v 2.0
Hospitals data linkage results by state NIHSI v 2.0 hospitals data were created by combining the NIHSI 1.0 data with the latest available 2019–20 hospitals data supplied to AIHW. The data were linked using a probabilistic linkage method. This is a method of linking using non-unique identifiers (e.g., name, date of birth, address) to establish weights which represent the likelihood that two records belong to the same person. These weights are used to inform matches and non-matches and include clerical review for a selected ‘grey area’ in between based on cut-off weights. These cut-off weights are selected based on the results of sample-based clerical reviews. For this project, high cut-off weights were chosen to ensure that estimated Positive Predictive Value (PPV) chosen as the linkage accuracy indicator is 98% or above. The PPV is calculated as percentage of the number of true positive links to the sum of true positive and false positive links. Percentage Linked The total number of records in the data set (cohort) were linked to the spine. This means for example, if we linked 85 records out of 100 records provided, the percentage of linkage rate is 85 per cent. The linkage results below reflect more about the linkage process that was undertaken using the latest available 2019–20 hospital data (referred to as a change file in the table). The table below provides the NIHSI v 2.0 linkage rates of the full data extract in the NIHSI data asset by combining NIHSI v 1.0 data with NIHSI 2019–20 change file data (see columns below).
a. Pre-linked data are the identifiers provided by the jurisdictional linkage nodes at person level. The linkage nodes pre-link the episodes, so the individuals with multiple admissions or ED presentations appear only once in the identifiers provided to AIHW. These individuals are represented as multiple records in non-pre-linked extracts, and there is no indications that these records belong to the same individuals. AIHW Data Integration Services Centre (DISC) determines this as part of the probabilistic linkage process. Therefore, the number of records in NSW reflects the number of episodes, but number of records in Qld extract reflects the number of individuals. b. Only Admitted Patients and ED presentations from Victoria were pre-linked. Non-admitted 2016–19 patients' identifiers were supplied separately and were not pre-linked. A significant proportion of these records (661,543) did not have names, only 44.4% of records in this group were linked. c. Percentage of all individuals or of all individuals with the identifiers in that data set that were linked to the National Map. The AIHW conducted a program of testing and validation to ensure the integrity and quality of NIHSI v. 2.0. These checks included:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Coherence: | Data held in the NIHSI v 2.0 are based on data held in the relevant AIHW national data collections. Differences in scope and methodologies used to create the NIHSI v 2.0 should be considered when comparing outputs with other sources. Demographic data, such as usual residence and/or Indigenous status, held in the NIHSI v 2.0 may vary across source data collections. Work is underway to understand the differences in the reporting of demographics within the Asset to be able to advise analysts. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Source and reference attributes | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Submitting organisation: | Australian Institute of Health and Welfare | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Origin: | Australian Institute of Health and Welfare | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Relational attributes | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related metadata references: | Supersedes National Integrated Health Service Information Analysis Asset (NIHSI AA) version 1.0 AIHW Data Quality Statements, Superseded 21/03/2024 See also COVID-19 Register; Quality Statement AIHW Data Quality Statements, Standard 07/03/2024 |