Sentiment Analysis of Patient- and Family-Related Sepsis Events: Exploratory Study (Preprint)
Bibliographic record
Abstract
BACKGROUND Despite the life-threatening nature of sepsis, little is known about the emotional experiences of patients and their families during sepsis events. We conducted a sentiment analysis pertaining to sepsis incidents involving patients and families, leveraging textual data retrieved from a publicly available blog post disseminated by the Centers for Disease Control and Prevention (CDC). OBJECTIVE This investigation involved a sentiment analysis of patient- and family-related sepsis events, leveraging text responses sourced from a publicly accessible blog post disseminated by the CDC. Driven by the imperative to elucidate the emotional dynamics encountered by patients and their families throughout sepsis incidents, the overarching aims centered on elucidating the emotional ramifications of sepsis on both patients and their families and discerning potential avenues for enhancing the quality of sepsis care. METHODS The research used a cross-sectional data mining methodology to investigate the sentiments and emotional aspects linked to sepsis, using a data set sourced from the CDC, which encompasses 170 responses from both patients and caregivers, spanning the period between September 2014 and September 2020. This investigation used the National Research Council Canada Emotion Lexicon for sentiment analysis, coupled with a combination of manual and automated techniques to extract salient features from textual responses. The study used negative binomial least absolute shrinkage and selection operator regressions to ascertain significant textual features that correlated with specific emotional states. Moreover, the visualization of Plutchik’s Wheel of Emotions facilitated the discernment of prevailing emotions within the data set. RESULTS The results showed that patients and their families experienced a range of emotions during sepsis events, including fear, anxiety, sadness, and gratitude. Our analyses revealed an estimated incidence rate ratio (IRR) of 1.35 for fear-related words and a 1.51 IRR for sadness-related words when mentioning “hospital” in sepsis-related experiences. Similarly, mentions of “intensive care unit” were associated with an average occurrence of 12.3 fear-related words and 10.8 sadness-related words. Surviving patients’ experiences had an estimated 1.15 IRR for joy-related words, contrasting with discussions around organ failure, which were associated with multiple negative emotions including disgust, anger, fear, and sadness. Furthermore, mentions of “death” were linked to more fear and anger words but fewer joy-related words. Conversely, longer timelines in sepsis events were associated with more joy-related words and fewer fear-related words, potentially indicating improved emotional adaptation over time. CONCLUSIONS The study’s outcomes underscore the imperative for health care providers to integrate emotional support alongside medical interventions for patients and families affected by sepsis, emphasizing the emotional toll incurred and highlighting the necessity of acknowledgment and resolution, advocating for the use of sentiment analysis as a means to tailor personalized emotional aid, and thereby potentially augmenting both patient and family welfare and overall outcomes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".