The way we use social media can give signs of how we are feeling. In this study, we will research whether summary data about tweets from the social media site Twitter can be used to know when the population needs more NHS mental health support, and how this changed throughout the COVID-19 pandemic. The tweets are all converted into summary scores so the original text will not be seen by researchers.
We will use information from the Avon Longitudinal Study of Parents and Children, such as scores of how positive or negative words extracted from participants' tweets are, data which were collected with participants' consent. It will be linked through UK LLC to their use of the NHS for mental health support, so we can find out whether Twitter was able to give early warning signs that the need for mental health care was changing for people in the UK.
Using the UK LLC dataset means that we will have high-quality data to test how accurate data extracted from Twitter is for measuring changes in NHS service demands. This is important because it will help us to develop accurate methods for understanding demand on the NHS in real time.
Nina Di Cara Nina.dicara@bristol.ac.uk