ATLAS network objective and rationale
The objective of the ATLAS network is to collate data to supplement the NNDSS and contribute to improved understanding of local, regional and national patterns of clinical care of STI and BBV among Aboriginal people.
Participating research partners
We invited five peak Aboriginal health organisations representing regional areas in four Australian jurisdictions to partner in this research. The four jurisdictions include Queensland, South Australia, New South Wales and Western Australia. Each organisation represents the views of multiple ACCHS within their geographical remit and agreed to act as ‘clinical hubs’ for this research. The five clinical hubs are Apunipima Cape York Health Council in northern Queensland; the Institute of Urban Indigenous Health (IUIH) in south-eastern Queensland; the Aboriginal Health and Medical Research Council of New South Wales; the Aboriginal Health Council of South Australia and the Kimberley Aboriginal Medical Services (KAMS) in Western Australia (Figure 1). These hubs were chosen based on convenience, geographic location and existing collaborative relationships with the research team. Each clinical hub is committed to oversight of the ATLAS network and other research generated as part of the study. It is our intention that the network expand over time.
Site engagement
The research team has committed to an extensive and ongoing community and site engagement process, which commenced while the funding proposal was being developed. Executive staff, sexual health/population health specialist staff and other management from the five clinical hubs and individual ACCHS were consulted about their organisation’s participation in the ATLAS network. Each ACCHS’s governing Board approved participation in the network.
Research governance and ethics
Formal approvals have been obtained from six Human Research Ethics Committee (HRECs) and regional research governance groups to date, including three Aboriginal-specific HRECs: the Aboriginal Health Research Ethics Committee in South Australia (EC00185, approval 04-17-732); the Aboriginal Health and Medical Research Ethics Committee (EC00342, approval 1300/17); and the Western Australian Aboriginal Health Ethics Committee (EC00292, approval 805, following approval from the Kimberley Aboriginal Health Planning Forum’s Research Subcommittee). Permissions were also obtained from the Far North Queensland Human Research Ethics Committee (as Queensland Health provides much of the primary care services in Aboriginal communities where the Apunipima Cape York Health Council is also active—EC00157, approval HREC/17/QCH/102–1173) and the Flinders University Social and Behavioural Research Ethics Committee (as the Lead Investigator’s affiliated institution—EC00194, approval OH-00142).
A Clinical Hub Reference Group, consisting of members from each clinical hub, provides critical oversight of and input to all research activities.
Operation of the ATLAS network
The ATLAS network routinely collects clinical data from participating ACCHS via their electronic medical record (EMR). ACCHS in the network contribute deidentified patient records relating to clinical care (testing, treatment and management) for in-scope STI and BBV either directly from the ACCHS’s EMR or via the third-party data extraction tool GRHANITEtm. EMR generally do not have the capacity to extract and analyse population-level STI and BBV data internally, sufficient to evaluate clinical practice and inform Continuous Quality Improvement (CQI). The ATLAS network provides this capacity.
Most EMR require the use of third-party software to perform the extraction and delivery of relevant data in a standardised manner. We explored several different commercially-available data extraction tools—including software developed by the Improvement Foundation (12) and Pen CS (13)—before opting for the University of Melbourne’s GRHANITEtm software (14). GRHANITE offered significant advantages over other software products including; compatibility with most of the EMR used in the ACCHS sector, capacity to deliver line-listed data, a substantial history of use by other sexual health research and surveillance projects (15-17), and best-practice approaches to patient deidentification and data encryption (9, 18).
For our network, the hash-based deidentification process was a key feature as it is thought to be more secure than the Australian Government’s SLK581 (19), retaining no element of personal information in the hashed identifier, yet still facilitates person-based linkage across the ATLAS network irrespective of the clinic from which the data originate. Similarly, the GRHANITE hashed identifier also can facilitate linkage to other projects and/or health services using GRHANITE. Moreover, the highly automated data collection process reduces workload for the research team once the surveillance infrastructure is established and enables further automation of the analysis and reporting system.
The initial data extraction from an ATLAS site collects data from 1 January 2016 to the date of extraction, followed by ongoing, regular (ideally, weekly) extractions. Deidentified EMR data are transferred to the ATLAS project’s secure databank located at the South Australian Health and Medical Research Institute (SAHMRI) and cleaned and stored for analysis. The ATLAS data processes utilise a custom-built SQL server accessing R Studio and Stata scripts interfacing with MS Word to produce standardised analyses for all sites participating in the surveillance network.
Data analysis
The STI currently included in the ATLAS project are chlamydia, gonorrhoea, trichomonas, syphilis and HIV. The BBV currently included in ATLAS are hepatitis B and hepatitis C (in addition to HIV).
An initial suite of 12 STI and BBV performance measures have been developed for the ATLAS network’s surveillance reports, and all are reflective of national clinical guidelines (6, 7) and having a strong basis in previous research in the sector (20-22)(see Table 1). Refinement of these performance measures, including their fit with any current reporting and their application to clinical practice, is an iterative and ongoing process conducted in collaboration with clinical hubs and services contributing to the ATLAS project. Development of these performance measures also incorporated input from related research projects involving members of the investigator group, to ensure that the needs of all projects working with the same ACCHS could be met by a singular data collection, analysis and reporting process. Baseline reports addressing the 12 performance measures, including sub-analyses and disaggregations, were calculated and drafts provided to clinical hub/health service staff for feedback before finalisation. Suggestions for improvements were immediately incorporated into our analysis and reporting templates where feasible and we continue to refine our processes based on ongoing feedback from sites as the surveillance network matures.
Table 1: ATLAS surveillance reporting Performance Measures
Performance measures
|
Definition
|
|
Proportion of clients tested for STIs (CT, NG, TV, syphilis and HIV) during the reporting period
|
|
Proportion of current clients tested for STIs at least once in a 12-month period
|
- Unique STI test positivity
|
Proportion of clients with at least one positive STI test in a 12-month period
|
- Completeness of STI Testing
|
Proportion of clients with a positive CT and/or NG and/or TV result also tested for syphilis and HIV within 30 days of date of initial specimen collection
|
|
Time (days) from date of positive STI (CT, NG, TV) test to date of treatment
|
|
Proportion of clients retested at approximately three months (60 to 120 days) following treatment for an initial positive STI (CT/NG/TV) result
|
- STI Repeat Positivity Rate
|
Proportion of clients retested at approximately three months (60 to 120 days) after treatment for an initial positive CT/NG result and who retested positive for CT/NG at this time
|
- Hepatitis B Virus (HBV) Testing Rate
|
Proportion of clients receiving an HBV test. Among those testing negative, the proportion who subsequently received one or more dose of vaccination.
|
- Hepatitis C (HCV) Testing Rate
|
Proportion of clients receiving an HCV antibody test and among those testing positive, the proportion then tested for RNA or viral load.
|
|
Proportion of HCV RNA positive clients prescribed Direct Acting Antiviral (DAA) treatment
|
- HCV Sustained Virological Response
|
Proportion of clients who, after having been prescribed DAA treatment, achieve an undetectable viral load
|
|
Proportion of female clients screened for human papillomavirus in line with national guidelines
|
Clients: all patients aged 15 years or older attending the health service for a clinical encounter or consultation with a medical doctor, nurse or Aboriginal health practitioner.
Abbreviations: CT Chlamydia trachomatis; DAA Direct Acting Antiviral; HBV Hepatitis B Virus; HCV Hepatitis C Virus; HIV Human Immunodeficiency Virus; HPV Human Papillomavirus; NG Neisseria gonorrhoeae; STI Sexually Transmissible Infection; TV Trichomonas vaginalis