Ms Priyanka Pillai1,2, Dr Kristal Spreadborough1
1Melbourne Data Analytics Platform (MDAP), The University of Melbourne, Parkville, Australia, 2The Peter Doherty Institute for Infection and Immunity, Parkville, Australia
Healthcare practitioners utilise electronic medical record (EMR) systems which collect both structured and free text data about their patients’ health and well-being. Whilst the analysis of structured data is relative straightforward, a method known as Natural Language Processing (NLP) shows increasing promise for the analysis of unstructured data. NLP is a processing mechanism whereby computers can understand human text and speech terms. NLP has the potential to harness valuable insights from data buried in the clinical free text that is routinely lost for reasons including additional resource implications and concern about increased privacy risk (compared to the analysis of structured clinical data).
There is literature to indicate that privacy concerns may impede NLP of unstructured clinical data for secondary use, particularly as such data is likely to include particularly sensitive information and qualitative clinical reflections. As the potential of NLP processing in this field continues to grow, it will be particularly important to be able to gauge public sentiment in relation to NLP processing of clinical free-text data in a robust, rapid and repeatable manner.
This paper presents a bioethics research protocol to engage with members of the public to assess public acceptability of secondary processing of their clinical free-text data, with a focus on NLP. We have adapted the Theoretical Framework of Acceptability (TFA) designed for clinical interventions to our empirical bioethics research and identified six key indicators of acceptability – ethicality, affective attitude, burden, value and benefits, perceived effectiveness, and coherence.
- A/Prof Mark Taylor, Melbourne Law School, The University of Melbourne, Parkville, VIC 3010, Australia
- Dr Megan Prictor, Melbourne Law School, The University of Melbourne, Parkville, VIC 3010, Australia
- Tess Whitton, Law PhD Student, Melbourne Law School, The University of Melbourne, Parkville, VIC 3010, Australia
- Minna Paltiel, Law PhD Student, Melbourne Law School, The University of Melbourne, Parkville, VIC 3010, Australia
- Kim Doyle, Melbourne Data Analytics Platform (MDAP), The University of Melbourne, Parkville, VIC 3010, Australia
- Geordie Zhang, Melbourne Data Analytics Platform (MDAP), The University of Melbourne, Parkville, VIC 3010, Australia
Priyanka Pillai has a background in bioinformatics and works as an academic specialist at the University of Melbourne. Priyanka works as a Research Data Specialist for the Melbourne Data Analytics Platform (MDAP) and as a Health Informatics Specialist for the Australian Partnership for Preparedness Research on Infectious Disease Emergencies (APPRISE) Centre of Excellence (CRE) based at the Peter Doherty Institute for Infection and Immunity in Melbourne, Australia.
Kristal Spreadborough is a Research Data Specialist at the Melbourne Data Analytics Platform. Kristal’s research interests cut across the fields of data, digital and data ethics, music, psychology and semiotics.