Using Data Integration to Improve Health and Welfare Insights
Abstract
:1. Introduction
- Broader level reporting and analysis—e.g., whole-population and national data;
- Addressing key data gaps by connecting content datasets that relate to a single entity;
- Analysis and reporting of rare or sensitive issues and events;
- Analysis of pathways taken through health and welfare systems and to understand the experience over a person’s life course;
- Identification of specific population groups in broader administrative datasets—e.g., migrants or veterans.
- Participation in Organization for Economic Co-operation and Development (OECD) activities, including the Working Party on Health Statistics;
- Longstanding involvement with the World Health Organization, through the WHO Family of International Classifications Network (WHO-FIC), and as the designated Australian Collaborating Center (ACC) for the WHO-FIC;
- Collaboration with the Canadian Institute for Health Information, through sharing and comparing approaches, and participating in secondments across the agencies;
- Membership with the National Initiative Network, a collective that shares experiences in developing stronger frameworks to promote secondary use of health and wellbeing data;
- Partnership with The Five Eyes research collective, with a particular focus on international comparisons of data about veterans;
- Contact with the United States of America National Center for Health Statistics, Statistics New Zealand, and the Commonwealth Fund.
2. AIHW Data Integration Projects
2.1. Dementia
- Better identification of people with dementia in Australia, leading to better coverage in reporting and more accurate understanding of disease prevalence, comorbidities, risk factors, and population groups with dementia.
- Developing an understanding of the course of disease over time for people with records of dementia including the potential to examine factors that affect the use of health and aged care services. Records of dementia may include a specific diagnosis, or recorded use of dementia-specific medications for diagnosis by proxy.
- Understanding the consequences of dementia diagnoses in many more aspects of a person’s life than previously reportable—e.g., on work and income, or receipt of welfare or disability support payments.
2.2. Disability
2.3. Health Service Use: Last Year of Life
2.4. Patient Experiences of Continuity of Care
2.5. Suicide
2.6. Veterans
- Overall causes of death and incidence of suicide for current serving and ex-serving ADF members [11];
- The welfare of ex-serving ADF members, from analysis of several topics including housing, social support, education and skills, employment, and income and finance;
- The use of healthcare services by ex-serving ADF members;
- Use of subsidized prescription medication by ex-serving ADF members;
- Health status, risk factors, and health conditions [12].
3. The AIHW Data Integration Process
3.1. Governance and Approvals
- Safe Projects—Is the use of the data appropriate (legal, moral, and ethical)?
- Safe Users—Can the users be trusted to use it in an appropriate manner?
- Safe Data—Is there a disclosure risk in releasing the data itself?
- Safe Settings—Does the access facility prevent unauthorized use?
- Safe Output—Are the statistical results non-disclosive [21]?
3.2. Strategic Partnerships
3.3. Quality Data
- Statistical risk—Managing statistical risks, which can occur at all stages and levels in the statistical production cycle, is key to maintaining data quality. To minimize statistical risk, the QMF provides clear definitions of the risks to data quality, as well as their significance (major, medium or minor), and provides guidance on developing strategies for their management.
- Project management—All statistical projects must complete a risk assessment at the planning stage. The project brief lists major risks, and any risks already realized are elevated to issues. Strategies for mitigation and management must be included. The risk assessment feeds into the design of the quality assurance and data validation strategies for each project. Risks are reviewed regularly throughout the project’s lifecycle.
- Quality assurance (QA)—QA strategies are particularly useful for identifying medium-level statistical risks and quality issues. They give a more detailed view of the factors impacting risks and data quality, often from a process perspective. The QMF provides context, generic tools, and a broad operational model to assist with the design of consistent QA strategies. It uses a set of generic gates to improve the early detection of errors or flaws in production processes. It also defines the roles and responsibilities for managing quality and performance measures to facilitate quality gate assessments.
- Data validation—Data validation processes present the last opportunity to detect, resolve, and treat important errors before the data are released to clients. Validation also enables anomalous data that are correct to be identified and explained. The QMF provides templates, guidance, and explanatory notes to assist with data validation work.
- Reference models—The QMF is based on reference models that integrate critical project management activities into the statistical production process, provide guidance on quality assurance and data validation work, and define the roles and responsibilities of stakeholders across project phases. These models can also be used to benchmark, monitor, understand, and streamline production processes, improving responsiveness and capability into the future.
3.4. Data Integration
- The separation principle means that no one working with the data can view both the linking (identifying) information (such as name, address, date of birth) together with the merged analysis (content) data (such as clinical information, health service, or medication usage) in an integrated dataset. Under the separation principle, data integration is performed in three stages—separation, linkage, and merging. Each stage has a separate domain within the specific project in the DISC. Each domain is accessible only by staff holding the specified role, and staff members can only perform one role in each project.
- Linkage is done on datasets containing essential data items only,
- Sophisticated probabilistic data linkage methodology is used to achieve the best possible linkage results. The linkage is performed using linkage software developed by the AIHW.
- Output is appropriately confidentialized before it is made available to researchers, in accordance with appropriate legislation and the requirements of data custodians.
4. Opportunities and Challenges
- Enhanced analysis—the research potential of integrated datasets is greater than of those based on a singular source. Integrated data have a broader coverage of topics and provide greater potential to examine interrelationships between topics.
- Cleaner data—the combination of data from different sources enables the development of improved data checks that can enhance the quality of the separate data sources. This can be achieved through the development of data collection standards or definitions of data items that relate to standard classifications.
- Cost effectiveness—linking data collected for other purposes is far cheaper than obtaining similar data through surveys and longitudinal studies. Reuse of existing administrative data greatly reduces the costs associated with both provision and collection.
- Improved coverage—linked datasets can represent a large sample, allowing broader-level reporting that is not possible from individual survey data. Use of integrated data can assist to address issues associated with small numbers. This both protects the privacy of individuals and enhances analysis and reporting of sensitive or rare events.
- Identification of target groups—linking information about target groups, e.g., migrants or veterans, creates broader datasets that support the analysis of wellbeing and identification of risk factors for these groups, without the requirement to ask for this detailed information in administrative datasets.
- Longitudinal analysis—integrating datasets over time can allow the analysis of pathways through health and welfare systems or over the life course of an individual or cohort.
- Coordination of the large and complex data integration landscape, involving data assets from all levels of government.
- Complex governance arrangements—including understanding the implications of relevant legislation, policies, and ethics. This requires a considerable amount of time and documentation.
- Managing liaison and approvals across multiple data custodians, especially across different levels of government. Integration projects drawing data from multiple sources typically require approvals from multiple ethics committees or custodians.
- Building community trust and engagement.
- Methodological challenges, where weighting practices may be required to ensure appropriate representativeness of the data.
- Data inconsistencies across input data sources. Data used in integration projects are often collected for service delivery or administrative purposes. They may have different definitions, concepts, specifications, coding, classifications, standards, and quality across sources.
- The need to quickly develop expertise in new and complex data models, as well as new approaches to analysis. As demand continues to grow for accessible and large-scale linked data assets such as the NIHSI and NDDA, the AIHW is responding to more complex, cross-sector research questions.
- Forming new and strengthening existing partnerships across all levels of government, to promote access to and use of data assets, as well as sharing of expertise.
- Promoting processes to safely share data for integration in national data assets that allows richer, deeper analysis of populations of interest. An example of this is the addition of population flags to national data assets, as in the AIHW veterans’ analysis work program.
- Continuous improvement and innovation in data collection practices, including opportunities to harmonize the way data on topics of interest are defined and collected.
- Supporting the development of governance frameworks that facilitate data integration involving data assets from all levels of government, while maintaining the privacy and confidentiality of data about individuals, as well as meeting the specific requirements of data custodians for access to and use of their data.
Funding
Institutional Review Board Statement
Informed Consent Statement
Acknowledgments
Conflicts of Interest
References
- Australian Government: Australian Institute of Health and Welfare (AIHW). Available online: https://www.aihw.gov.au/about-us (accessed on 6 July 2021).
- AIHW. Available online: https://www.aihw.gov.au/our-services/data-linkage/our-secure-linkage-environment (accessed on 6 July 2021).
- Australian Government. Available online: https://toolkit.data.gov.au/Data_Integration_-_Accredited_Integrating_Authorities.html (accessed on 6 July 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports/dementia/predicting-early-dementia-using-medicare-claims/contents/summary (accessed on 30 October 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports-data/health-welfare-overview/health-care-quality-performance/coordination-of-health-care (accessed on 26 August 2021).
- Australian Government: Department of Health. Primary Health Networks. Available online: https://www.health.gov.au/initiatives-and-programs/phn (accessed on 26 August 2021).
- Coordination of Health Care for Patients Aged 45 and Over by Primary Health Networks, Summary—Australian Institute of Health and Welfare. Available online: aihw.gov.au (accessed on 26 August 2021).
- AIHW. Available online: https://www.aihw.gov.au/suicide-self-harm-monitoring/data/behaviours-risk-factors/social-factors-suicide (accessed on 26 August 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports-data/population-groups/veterans/overview (accessed on 6 July 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports/veterans/a-profile-of-australias-veterans-2018/summary (accessed on 6 July 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports/veterans/serving-and-ex-serving-adf-suicide-monitoring-2021 (accessed on 30 October 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports-data/population-groups/veterans/reports (accessed on 6 July 2021).
- AIHW. Available online: https://www.aihw.gov.au/reports/veterans/adf-members-population-characteristics-2019 (accessed on 30 October 2021).
- Australian Government. Available online: https://www.legislation.gov.au/Details/C2020C00237 (accessed on 19 August 2021).
- Australian Government: Office of the Australian Information Commissioner. Available online: https://www.oaic.gov.au/privacy/australian-privacy-principles (accessed on 19 August 2021).
- AIHW. Available online: https://www.aihw.gov.au/about-our-data/data-governance (accessed on 19 August 2021).
- AIHW. Available online: https://www.aihw.gov.au/about-us/privacy-policy/aihw-privacy-policy (accessed on 30 August 2021).
- Australian Government. Available online: https://www.legislation.gov.au/Series/C2004A03450 (accessed on 26 August 2021).
- AIHW. Available online: https://www.legislation.gov.au/Details/F2018L00317 (accessed on 30 August 2021).
- Five Safes. Available online: www.fivesafes.org (accessed on 6 July 2021).
- AIHW. Available online: https://www.aihw.gov.au/about-our-data/data-governance/the-five-safes-framework (accessed on 6 July 2021).
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Jensen, L.R. Using Data Integration to Improve Health and Welfare Insights. Int. J. Environ. Res. Public Health 2022, 19, 836. https://doi.org/10.3390/ijerph19020836
Jensen LR. Using Data Integration to Improve Health and Welfare Insights. International Journal of Environmental Research and Public Health. 2022; 19(2):836. https://doi.org/10.3390/ijerph19020836
Chicago/Turabian StyleJensen, Linda R. 2022. "Using Data Integration to Improve Health and Welfare Insights" International Journal of Environmental Research and Public Health 19, no. 2: 836. https://doi.org/10.3390/ijerph19020836
APA StyleJensen, L. R. (2022). Using Data Integration to Improve Health and Welfare Insights. International Journal of Environmental Research and Public Health, 19(2), 836. https://doi.org/10.3390/ijerph19020836