Data is the primary resource - the life blood - that drives evidence-based health and medical (H&M) research. Traditionally, data collected explicitly for a clinical trial, designed to answer specific questions on behavioural or bio-medical questions, served as the primary research data for H&M researchers. This practice started when administrative health data was not readily available in computable electronic format. However, in the current digital health environment, administrative health data collected and aggregated in electronic form has become a veritable gold mine for stakeholders in the healthcare environment. As a consequence, administrative data is increasingly becoming the primary data source for many H&M researchers.
The purpose of collecting administrative health data are manifold:
- to manage the health care of individual consumers and ensure continuity of care
- to streamline health care service delivery at all levels
- to manage and administer hospital and health care service delivery
- to keep track of healthcare costs including billing for goods and services delivered
- to inform health system policy
- to ensure reliable and consistently high quality of care.
In Flying Blind Report Volume 1, we highlighted the acute level of fragmentation in clinical and administrative data that is routinely collected at all points of service, including primary care, private and public hospital care, emergency care, aged care, etc. Most of these electronic datasets remain in silos and are aggregated and reported to public and private health funders, state and federal governments and other regulatory agencies and research agencies such as APRA and AIHW.
Primary and Secondary Uses of Health Data
While the primary purpose of administrative data collection is focussed on various aspects of healthcare service delivery, informed analysis of the detailed information contained within these administrative and clinical datasets can offer valuable insights that go far beyond the original intent for which the data were collected. Data, be it research data or administrative data, when used for purposes that go beyond the original intent, is often referred to as secondary use of primary data.
As Australian health data is well coded and standardised, researchers and government agencies can add significant value by harnessing these administrative data assets to inform policy and quality of care. In this context administrative health data sets become primary source for many H&M research projects. In addition, when fragmented data sets are de-identified and linked, the entire health sector, including healthcare consumers can derive significant benefits.
In addition to primary data sets, researchers and policy makers use additional sources of secondary data that could be external to the system. Secondary data refers to data that was collected by another agency for some other purpose. Common sources of secondary data for health and social science include censuses, information collected by government departments, organizational records and data that was originally collected for other research purposes.
The researcher’s journey discussed in an earlier post is indeed a by-product of all the confusion around linking primary and secondary data and secondary use of primary data.
The recent Productivity Commission Draft Report on Data availability and use acknowledges the “extraordinary growth in data generation and usability combined with computing power, Internet connectivity and algorithms, have enabled a kaleidoscope of new business models, products and insights … and Individuals, businesses, governments and the broader community have all benefited from these changes."
The Report also recognizes that “marginal changes to existing structures and legislation will not suffice. The Commission is proposing reforms to data availability and use, aimed at moving from a system based on risk aversion and avoidance, to one based on transparency and confidence in data processes.”
In the context of health research, while ethics committees and data custodians ensure that all individual data is de-identified and anonymised before it is used by researchers, as identified in the PC Draft Report, “frameworks and protections developed for data collection and access prior to sweeping digitization now need reform. This is a global phenomenon and Australia, to its detriment, is not yet participating. Lack of trust and numerous barriers to sharing and releasing data are stymieing the use and value of Australia’s data.”
What we now need is a transparent and unambiguous governance framework that is designed and articulated to address the ethical concerns around the many uses of primary and secondary data as well as the issues around multiple uses of linked datasets. These ethics and data custodian related issues need to be addressed both boldly and cautiously, keeping in mind the significant benefits that can be gained by the healthcare sector at large when anonymised de-identified linked datasets are used by researchers in highly secure environments to inform clinical and health research and healthcare policy.
With clearly articulated data management and governance framework, privacy policies and standardized state-of-the art de-identification techniques, Australia is well positioned to exploit the high quality digital data to deliver targeted high quality healthcare that include the full complement of services starting with preventive healthcare, treatment services and wellness and health management strategies that can benefit consumers as well as service providers and address the many concerns of funders, payers and policy makers.