Dr Lamiece HassanDr Hassan was selected as an HDR UK Fellow to support and build her expertise in health data science. Based at the University of Manchester, she leads digital health research across a range of health informatics projects. Dr Hassan’s HDR UK funded project will test new ways of applying natural language processing methods to analyse social media data to assess public opinion on health and care using a mixed methods approach.

 

“The Fellowship has been a brilliant opportunity for me to change track in my career; it’s opened doors and is providing me with skills and training needed to navigate technical and ethical challenges to further my research into patient experience,” said Dr Hassan.

How is the HDR UK fellowship supporting you to be a future leader?

This three-year fellowship is giving me the time, training, and networks to develop my health data science skills and collaborate with world-leading experts to bring new methods to my work. In particular, this includes techniques for analysing social media data about experiences and opinions relevant to healthcare, data analysis and digital technologies.

My current projects aim to find better ways of channelling patient opinions and experiences to shape health services for the benefit of patients. I’ll be investigating what kinds of insights can be picked up from online words, text and conversations posted publicly on social media platforms.

How will you go about this?

I’m looking to learn and trial a technique called Natural Language Processing (NLP) to analyse naturally occurring conversations, particularly on social media. It’s a way of harnessing patient experience in a different way by deriving meaning from text. Often a patient may have an adverse reaction to a drug or be unhappy with a health service, but whilst they may not tell their doctor, they may write about it on Twitter or Facebook.

One area that interests me is understanding the public’s perception of the NHS working with commercial companies, such as a pharmaceutical or software company. If you can understand the public’s concerns you can look to address them and set standards for working in trustworthy ways.

The plan is to use NLP alongside more traditional ways of understanding patient opinions and experience, to spot any pitfalls or limitations, and find ways that these insights might shape future care, policy and services.

How else do you gain insights?

I am also interested in the free text found in notes, letters and reports within medical records as a way of providing a richer narrative to patient experience. Analysing free text comes with a host of challenges as it can contain sensitive and personal data that may need to be removed, so there are ethical, confidential and security issues to contend with. This is where my public engagement background comes in and I am collaborating with other researchers to better understand the public’s views on this.

What impact has your research had?

Recently, Dr Elizabeth Ford of Brighton and Sussex Medical School invited me to help run a Citizens Jury: a deep dive into public opinion on whether it’s acceptable to use free text in medical records for research. At first people didn’t necessarily know what their medical record looked like, never mind the difference between structured and unstructured data. After understanding the process, and considering the pros and cons of access, everyone fairly or strongly supported access to the data provided the right safeguards were in place. Our next step is to finalise a series of recommendations for the governance of free text research in future, that will hopefully influence researchers and data custodians.