JMIR Publications

ABSTRACT

Background:

The use of social media in cancer research, patient support, and information sharing has been well documented

Objective:

Using retinoblastoma as a model, we utilise the abundance of information provided from social media to understand patient’s treatment seeking behaviour and barriers, as well as investigate its application in research and epidemiology.

Methods:

Posts on retinoblastoma were extracted from 6 social media sites. We trained 3 Bidirectional Encoder Representations from Transformers models (BERT) to identify relevance and conducted sentiment analysis with a pre-trained BERT model. The Hierarchical Dirichlet Process was trained to identify topics. We enriched user profiles with OpenStreetMap-based geotagging and coreNLP-based occupation tagging. Retinoblastoma outcomes were obtained from systemic review and meta-analysis, which covered articles published between January 1, 1981, and October 8, 2021.

Results:

2382511 entries in total were extracted from 6 social media sites. Social media channels had different focuses, with retinoblastoma campaigns the most prevalent topic on Twitter, while retinoblastoma treatment was mainly discussed on YouTube and Weibo. Weixin users focused on retinoblastoma presentation and diagnosis. However, most of the information dissemination and discussion originated from North America and Western Europe. Regions with more reluctance towards enucleation were associated with poorer survival (β=-0.999, CI=-1.571– -0.428, P=.0011) and globe salvage rate (β=-0.545, CI=-1.002–-0.087, P=.0209). Active participation of clinical staff (β=-0.105, CI=-0.186–-0.024, P=.0111) and academia (β=-0.116, CI=-0.208–-0.252, P=.0126) correlated with lower enucleation hesitancy. Lack of disease understanding and the need for more support and counselling remained the most significant barrier to receiving treatment worldwide, as reflected by both intensity and number of tweets.

Conclusions:

Social media is extensively used for retinoblastoma education and support especially in North America and Western Europe. More positive sentiment on social media for retinoblastoma and treatment is associated with better survival. Targeted information dissemination by trained personnel should improve acceptance in vulnerable zones and outcomes. Clinical Trial: N.A.