Computational Sentiment Analysis of LVAD Support Group: Positivity Predominates

Abstract

Despite well-documented left ventricular assist device (LVAD) complications, the impact of these complications on the individual patient, as well as the overall sentiment and its effect on referral patterns in the patient community, is not fully understood. We sought to better understand patient attitudes towards LVAD therapy using a computational sentiment analysis approach. The publicly available LVAD community support forum mylvad.com was chosen for analysis. The posts, comments, and titles were parsed from the forum's html using a custom Python script (Version 3.6). Individual word frequency was computed and each word was classified as 'positive', 'negative', or 'neutral'. Data processing, sentiment determination, and analysis were performed with a binary dictionary package using R software (Version 3.6). After text transformation and cleaning, 66,821 unique words remained, including 4,623 (6.9%) with positive sentiment and 3,248 (4.8%) with negative sentiment. The net sentiment ratio (Number of Positive Words - Number of Negative Words) / (Number of Total Words) was 2.1%. Positive sentiment dominated the 20 most commonly used words (Fig. 1A). The odds ratio of non-neutral words (Number of Positive Words / Number of Negative Words) was 1.42, indicating a less obvious disparity in sentiment when expanding the analysis further than the top 20 words. Word cloud analysis of positive (green words) and negative (red words) sentiments is shown (Fig. 1B), indicating a more frequent use of "infection" (208 mentions) than other known LVAD complications such as stroke (29 mentions), bleeding (30 mentions), and thrombosis or clot (32 mentions). Positive sentiment dominates among the most frequently used words, yet this disparity decreases when considering the totality of words. The negative word "infection" is mentioned a disproportionate number of times compared to other often-cited LVAD complications. This may provide practitioners insight into predominant patient concerns.