Research by Cecilia Price and Leo Petersen-Khmelnitski
Secondary use of health data can help health systems to better understand and serve a population’s needs. Primary use of health data means using an individual’s health information to inform their care, for example, if a diagnostic test demonstrates an individual carries an antibiotic resistant strain of bacteria, this will dictate what medicine they are given. In contrast, secondary use of health data involves aggregating health data from various sources (electronic health records, wearable technologies, health insurance data and more) and using it to improve health and research and development, policymaking, care planning and safety monitoring. To continue with the earlier example, data on incidence of antibiotic resistance can be aggregated and used to monitor resistance on a population level and to make policies to tackle resistance. Here, we address where this data comes from, why secondary use of health data is important and the barriers to progress.
Where does the data come from?
The data used for secondary purposes comes from a multitude of sources. First, data from clinical settings can be used, this includes clinical trials data and data from electronic health records. Real-world data (data from outside the clinical setting) can also be used, for example, data from sickness and insurance claims records, from wearables and other devices. Significantly, this data can be anonymised and aggregated to generate insights on a population level.
Examples of secondary use of health data
Health data can be used from research to policy to implementation, and there are a number of examples of existing secondary uses of health data:
- COVID-19 symptom tracker apps collate surveillance data that can guide research and early diagnostics
- AMR surveillance data can aid early detection of resistant strains and outbreak investigations
- Health data from registries can be used for research, for example, researchers can request health data from danishhealthdata.com
- Aggregation of genomic data, such as that by the 1+ Million Genomes Project, can create a better understanding of disease and support precision medicine and precision public health as well as other innovation
What are the benefits of secondary use of health data?
The secondary use of health data creates a number of opportunities and benefits. First, secondary use of health data can improve health systems. In the face of limited resources, policymakers can use health data to prioritise resources to best meet the population’s needs.
Secondary use of health data can also facilitate the shift towards prevention, by enabling early diagnostic and surveillance systems; such prevention efforts are vital in the face of ageing populations and rising chronic disease.
Secondary use of health data can also improve the patient journey, as it enables remote monitoring and care via digital tools as well as early, personalised diagnostics and personalised care pathways.
The opportunity to leverage real-world data and for individuals to contribute their personal data means higher patient-public participation in the way that health is designed and delivered, which can ultimately improve care outcomes.
Lastly, secondary use of health data can enable innovation, by creating opportunities for new research, development of new medicines and technologies, and predictive modelling.
What are the challenges of using health data for secondary purposes?
First, data access, and specifically access to quality data, is a hurdle when using health data for secondary purposes: to use health data for secondary purposes like research and policymaking, researchers and policymakers must be able to access appropriate, high quality data. Linked to this, interoperability is a key challenge when it comes to creating a functional data ecosystem, as secondary use of health data requires the triangulation and integration of data from multiple sources. In addition, there are concerns around data sharing, privacy and governance, all of which must be inbuilt to any data ecosystem.