ABSTRACT
Background: Dementia misconceptions on Twitter can have detrimental or harmful effects. Machine learning (ML) models codeveloped with carers provide a method to identify these and help in evaluating awareness campaigns. Objective: This study aimed to develop an ML model to distinguish between misconceptions and neutral tweets and to develop, deploy, and evaluate an awareness campaign to tackle dementia misconceptions. Methods: Taking 1414 tweets rated by carers from our previous work, we built 4 ML models. Using a 5-fold cross-validation, we evaluated them and performed a further blind validation with carers for the best 2 ML models; from this blind validation, we selected the best model overall. We codeveloped an awareness campaign and collected pre-post campaign tweets (N=4880), classifying them with our model as misconceptions or not. We analyzed dementia tweets from the United Kingdom across the campaign period (N=7124) to investigate how current events influenced misconception prevalence during this time. Results: A random forest model best identified misconceptions with an accuracy of 82% from blind validation and found that 37% of the UK tweets (N=7124) about dementia across the campaign period were misconceptions. From this, we could track how the prevalence of misconceptions changed in response to top news stories in the United Kingdom. Misconceptions significantly rose around political topics and were highest (22/28, 79% of the dementia tweets) when there was controversy over the UK government allowing to continue hunting during the COVID-19 pandemic. After our campaign, there was no significant change in the prevalence of misconceptions. Conclusions: Through codevelopment with carers, we developed an accurate ML model to predict misconceptions in dementia tweets. Our awareness campaign was ineffective, but similar campaigns could be enhanced through ML to respond to current events that affect misconceptions in real time.
ABSTRACT
BACKGROUND: Dementia misconceptions on social media are common, with negative effects on people with the condition, their carers, and those who know them. This study codeveloped a thematic framework with carers to understand the forms these misconceptions take on Twitter. OBJECTIVE: The aim of this study is to identify and analyze types of dementia conversations on Twitter using participatory methods. METHODS: A total of 3 focus groups with dementia carers were held to develop a framework of dementia misconceptions based on their experiences. Dementia-related tweets were collected from Twitter's official application programming interface using neutral and negative search terms defined by the literature and by carers (N=48,211). A sample of these tweets was selected with equal numbers of neutral and negative words (n=1497), which was validated in individual ratings by carers. We then used the framework to analyze, in detail, a sample of carer-rated negative tweets (n=863). RESULTS: A total of 25.94% (12,507/48,211) of our tweet corpus contained negative search terms about dementia. The carers' framework had 3 negative and 3 neutral categories. Our thematic analysis of carer-rated negative tweets found 9 themes, including the use of weaponizing language to insult politicians (469/863, 54.3%), using dehumanizing or outdated words or statements about members of the public (n=143, 16.6%), unfounded claims about the cures or causes of dementia (n=11, 1.3%), or providing armchair diagnoses of dementia (n=21, 2.4%). CONCLUSIONS: This is the first study to use participatory methods to develop a framework that identifies dementia misconceptions on Twitter. We show that misconceptions and stigmatizing language are not rare. They manifest through minimizing and underestimating language. Web-based campaigns aiming to reduce discrimination and stigma about dementia could target those who use negative vocabulary and reduce the misconceptions that are being propagated, thus improving general awareness.