Data Collection

First data will be collected as the patient provides their permission for the A3BC/ARAD to contact them, via an affiliated Rheumatologist or health professional. Using a sophisticated laboratory information management system (LIMS) and online A3BC registry, a unique identifier will be generated for each participant which will be used to identify them in all correspondence with external parties that are not authorised to see personal health information. In this manner, only those with appropriate authority (HREC approved PIs, and who have signed a confidentiality agreement) will be able to link the person’s identity to their record if required. It is important that they have access to identifiable data as they will need to cross-match records from State Health EMR, pathology and radiology that will be linked in to the A3BC. From the first point of biospecimen collection, the unique A3BC participant identifier will be linked (by barcode and data linkage between systems) to every biospecimen, subsequent sample, article of documentation, and data entry.

Collection Schedule
Questionnaire data is collected and 0, 6, 12, 18 and 24 months, then 12-monthly for the life of the project. Data may be collected beyond these times where significant changes in disease (flares) or treatment status occur. Where participants change/switch therapies, the collection timeline will not refresh (unless specific to an approved project). This will provide suitable longitudinal analyses across the disease course, from each target disease group and sub-group of interest. Data-linked information from the datasets below will be sought at the same timepoints in order to qualify cross-dataset association analyses.

Minimum Dataset
In aligning with international best practice, and ensuring provision of high-quality longitudinal data (which randomized controlled trials cannot capture) to enable innovative research design and discovery, the A3BC has developed a minimum dataset. The A3BC Minimum Dataset (A3BC MD), was developed in consideration of current consensus by the OMERACT (Outcome Measures for Arthritis Clinical Trials) and collaborators regarding observational cohort study in RA studies and other rheumatic conditions . The A3BC will collect the A3BC MD elements through the complimentary collection of clinical information within the CRF, and patient information within questionnaires.

Pre-Analytical Variables
As part of its quality management of samples, the A3BC will collect pre-analytical variables on all blood and tissue biospecimens from collection to processing to storage. These include ischemia time, type of vacutainer, collection to processing time, centrifugation and long-term storage containers/temperature. The variables recorded are accordance with the Standard Preanalytical Coding for Biospecimens (version 2.0), developed by the International Society for Biological and Environmental Repositories (ISBER).

Medical Imaging Data
The A3BC will routinely collect histology images and radiology reporting for investigations such as X-ray, MRI, and ultrasound, from the participant’s physician. Histology images will be placed in the A3BC system, linked to case tissues, so applicants can preview tissue composition (via a secure temporary link) before proceeding to full application. As suggested by recent review, the incorporation of digital pathology into biobank quality assurance procedures, can reduce variability and subjectivity inherent to the pathology evaluations of biobanks. Notably, these images can both assist researchers in sample selection and be a resource in themselves. Low resolution radiology images linked to the participant may also be collected and stored if cost-effective, While, given the storage demands (high volume file sizes) and currently unknown researcher demand for use, the collection and storage of full resolution radiology images will be considered on a project basis.

Data Linkage

Data linkage visualisation and analysis activity will be conducted via state-of-the-art high-powered local/cloud computing laboratories. Of note, the Secure Unified Research Environment (SURE) was established with NSW Government and Australian Government National Collaborative Research Infrastructure Strategy (NCRIS, as part of the Population Health Research Network) funding support. Ethics-approved staff and researchers with access to visualisation, storage or analytics systems containing participant information will have a secure, trackable login.

Patient data is sourced through data linkage to the following:

  • Rheumatologist clinics for clinical outcomes and histories
  • Australian Rheumatology Association Database (ARAD)
  • Commonwealth health data (MBS, PBS, Cancer registry)
  • Longitudinal/Lifecourse data (CLARITY, NWAHS, etc)
  • Electronic medical records (EMR, Australia-wide local health districts/networks):
    • Statewide patient identifiers and national Individual Healthcare Identifier
    • Hospital encounter history from Emergency, Inpatients and Outpatients departments
    • Discharge summaries from public hospitals
    • Clinical observations (e.g. BMI, blood pressure)
    • Pathology and radiographic images and reports
    • Community Health Services episodes of care data
    • Admitted Patient Data Collection
    • Medications and immunisations
    • Allergies and adverse reactions
    • My Health Record information
    • State data linkage units
    • Cancer registries
    • Death registries

Ongoing patient-centric EMR data extraction will be managed by a State Health Department registered and accredited data extraction supplier. Commonwealth data linkage will occur by two methods. (1) Patients already recruited, or concurrently to the ARAD, which has approved Commonwealth linkage, will be reconsented and provide biospecimens. (2) Approval from the DHS for A3BC biobank-consented patients outside of the ARAD. To streamline the proposed data linkage scope, the A3BC are developing enduring data linkage processes through discussions with the NSW Statewide Biobank and the Commonwealth. Information between the historical ARAD and new A3BC will be coded and linked using study identifiers, as explained to participants during the A3BC consent process.

Data Analytics

The Australian Commission on Safety and Quality in Health Care (ACSQHC) has identified musculoskeletal conditions as one of the two top priority areas for developing clinical quality registries because of the very high cost and high disease burden of these conditions [1]. Firstly, the A3BC will enhance the scope of the ARAD by allowing existing data to be matched with biospecimens, support increased recruitment of clinicians and patients, and enhance clinical outcome data collection from clinicians, to better inform clinical policy and practice. Secondly, A3BC systems will support rheumatologist clinical quality audits/projects. While the ACSQHC recently commended groups providing patient-reported outcomes (PROs) on a six-monthly basis, the A3BC will take timely reporting to a new level, providing clinician dashboards with real-time analytics, not months later, but on demand [2]. The dashboarding will include a clinic-driven survey subset and link to electronic medical records for improved clinician-patient decision-making, satisfaction and treatment adherence at the point of care.

[1] The Australian Commission on Safety and Quality in Health Care. Prioritised list clinical domains for clinical quality registry development: ACSQHC; 2016.
[2] Thompson C, et al. Patient-reported outcome measures: an environmental scan of the Australian healthcare sector: ACSQHC; 2016.