Leveraging Online Patient Forums to Understand Breast Reduction Concerns: A Machine-Learning Analysis of 3078 Patient Questions Over 15 Years.
BACKGROUND: Patients are increasingly using social media and online forums to learn about plastic surgery, which can influence their expectations. Understanding patient concerns on these platforms will facilitate productive clinic discussion and ensure patients are receiving accurate, evidence-based information. OBJECTIVES: The aim of this study was to analyze breast reduction questions posted on RealSelf (Seattle, WA), an online plastic surgery forum. METHODS: The website www.realself.com/questions/breast-reduction was accessed on June 9, 2023. Posting date and poster self-reported location were extracted. Question header and text were manually reviewed. Questions were categorized by timing (preoperative vs postoperative) and topic. Regional and temporal trends were assessed. A machine-learning (ML) algorithm was applied to identify the top (most representative) preoperative and postoperative questions. RESULTS: In total, 3078 questions from August 2008 to May 2023 were analyzed. Questions most frequently originated from the southern United States (34.5%) and were asked preoperatively (58.4%). The most common question topics were postoperative care (24.9%), postoperative appearance/sensation (15.7%), and surgical logistics (10.2%). The distribution of topics varied significantly between location (P < .01), with topics such as insurance (P < .01) more likely to be asked in the south. CONCLUSIONS: This is the first study to leverage ML workflows to analyze a large volume of patient questions about breast reduction from an online plastic surgery forum. Analyzing patient questions on social media and online forums such as RealSelf with ML techniques can provide valuable insight into common concerns and informational gaps surrounding plastic surgery. Plastic surgeons should consider these results to guide patient conversations, combat misinformation, and facilitate deliverance of efficient care.
Duke Scholars
Published In
DOI
EISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Surveys and Questionnaires
- Surgery
- Social Media
- Patient Education as Topic
- Mammaplasty
- Machine Learning
- Humans
- Female
- 3202 Clinical sciences
- 11 Medical and Health Sciences
Citation
Published In
DOI
EISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Surveys and Questionnaires
- Surgery
- Social Media
- Patient Education as Topic
- Mammaplasty
- Machine Learning
- Humans
- Female
- 3202 Clinical sciences
- 11 Medical and Health Sciences