https://orcid.org/0000-0002-3042-6091
Figures
Abstract
Background
Taking part in a cancer clinical trial often represents a source of psychological distress and emotional activation among patients and their caregivers. Nowadays, social media platforms provide a space for these groups to freely express and share their emotional experiences.
Aims
We aimed to reveal the most prevalent basic and complex emotions and sentiments in the posts of the patients and caregivers contemplating clinical trials on Reddit. Additionally, we aimed to categorize the types of users and posts.
Methods
With the use of keywords referring to clinical trials, we searched for public posts on the subreddit ‘cancer’. R studio v. 4.1.2 (2021-11-01) and NRC Emotion Lexicon was used for analysis. Following the theoretical framework of Plutchik’s wheel of emotions, the analysis included: 8 basic emotions (anger, fear, anticipation, trust, surprise, sadness, joy, and disgust) and 4 types of complex emotions (primary, secondary, tertiary, and opposite dyads). We utilized the package ‘PyPlutchik’ to visualize the emotion wheels in Python 3.10.5.
Results
A total of 241 posts were included in the final database. User types (129 patients, 112 caregivers) and post types (142 expressed shared experience, 77 expressed advice, and 85 conveyed both) were identified. Both positive (N = 2557, M = .68) and negative (N = 2154, M = .57) sentiments were high. The most prevalent basic emotions were: fear (N = 1702, M = .45), sadness (N = 1494, M = .40), trust (N = 1470, M = .44), and anticipation (N = 1376, M = .37). The prevalence of complex/dyadic emotions and their interpretation is further discussed.
Conclusion
In this contribution, we identified and discussed prevalent emotions such as fear, sadness, optimism, hope, despair, and outrage that mirror the psychological state of users and affect the medical choices they make. The insights gained in our study contribute to the understanding of the barriers and reinforcers to participation in trials and can improve the ability of healthcare professionals to assist patients when confronted with this choice.
Citation: Chichua M, Filipponi C, Mazzoni D, Pravettoni G (2023) The emotional side of taking part in a cancer clinical trial. PLoS ONE 18(4): e0284268. https://doi.org/10.1371/journal.pone.0284268
Editor: Takeru Abe, Yokohama City University, JAPAN
Received: August 30, 2022; Accepted: March 27, 2023; Published: April 24, 2023
Copyright: © 2023 Chichua et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data was extracted from publicly available posts. The following URLs were used for downloading the posts: https://api.pushshift.io/reddit/search/comment?q=new*drug&subreddit=cancer&size=500 https://api.pushshift.io/reddit/search/comment?q=clinical*trial&subreddit=cancer&size=500 https://api.pushshift.io/reddit/search/comment?q=new*treatment&subreddit=cancer&size=500 https://api.pushshift.io/reddit/search/comment?q=experimental*trial&subreddit=cancer&size=500 https://api.pushshift.io/reddit/search/comment?q=experimental*drug&subreddit=cancer&size=500 https://api.pushshift.io/reddit/search/comment?q=experimental*treatment&subreddit=cancer&size=500.
Funding: This work was partially supported by the Italian Ministry of Health with Ricerca Corrente and 5x1000 funds.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Cancer patients can see participating in a clinical trial as an opportunity with several benefits, but it also represents a source of psychological distress and emotional activation [1]. For example, patients who are offered to participate in clinical trials testing the safety or efficacy of new therapies can experience fear of the unknown [2] and the fear that they will be considered only as one experimental subject rather than a person [1]. Concerns, such as uncertainty and lack of social support, may take a severe toll on their consent to enroll and their adherence afterward [3]. This is particularly relevant if we consider that cancer elicits special needs and makes patients and their caregivers more vulnerable [4]. Indeed, experiencing cancer and the related treatment significantly impacts the patients’ lives, which may lead patients and caregivers to an overwhelming emotional activation [5].
However, even if patients highly regard clinicians for their medical knowledge and skills, they may perceive some barriers to expressing their emotional experiences to the clinicians, such as the lack of time [6] and the physicians’ lack of ‘first-hand experience’ [7, 8]. For what concerns taking part in clinical trials, sharing the negative emotional experience with their oncologist might increase patients’ fear; they may fear hurting their relationship with their doctor or receiving worse treatment in the future due to sharing their doubts [9].
Finding a safe space for expressing and sharing these emotional experiences is thus fundamental, and nowadays, social media platforms present an opportunity for patients and caregivers. Emotional needs are one of the main reasons patients turn to health-related online social platforms [10] and online health-related communities provide users with the support they cannot get from their natural social network [7]. In this regard, in a recent systematic literature review [9], social support (including emotion and network support), emotional expression, and social comparison were identified as the main categories of health-related social media use. These categories point to the psychosocial needs of patients and their caregivers and the complementary role social media plays alongside traditional healthcare facilities.
Consequently, several researchers have turned to online social media platforms, utilizing various natural language processing (NLP) techniques to analyze user posts and comments. It has been suggested that exploration of such platforms may help improve traditional psychological assessments (e.g., European Organization for Research and Treatment of Cancer [EORTC] Quality of Life Questionnaire Core 30 [QLQ-C30] and McGill Pain Questionnaire (MPQ)) that are administered to patients in medical settings and thus, address the gaps that exist when using these traditional tools [11, 12]. For example, with the help of artificial intelligence and emotion analytics algorithms, it has been possible to identify the descriptive terms of pain that have been outdated yet still used in the questionnaire and, on the contrary, the terms that are more frequently used in daily life yet not included in the assessment tool [12]. Moreover, by utilizing topic analysis of social media posts, new areas of quality of life of breast cancer patients have been described [11]. These findings demonstrate the evolving nature of language and the utility of the NLP techniques that allow for ecological qualitative online data analysis.
NLP has also been applied to explore the emotional side of online discussions. A recent sentiment analysis [13] has shown that on an online breast cancer support group, the most active users used positive language significantly more frequently than the least active users. Moreover, negative words have been significantly higher for the latter group. The difference in sentiment was evident between users with different stages of cancer as well. Similarly, the text polarity classes such as positive, neutral and negative have been the focus of another recent study that successfully predicted the sentiment of patient-authored content on e-health forums using various machine learning algorithms [14]. The importance of further research in emotion and sentiment analysis has been outlined in these works.
Focusing primarily on emotional expressions and words used online by patients and caregivers when they discuss clinical trials is important as this topic is often the source of conflicting feelings. Even though the literature on the emotion analytics of online textual material is expanding, the emotional expressions and linguistic patterns of patients and caregivers discussing clinical trials still need to be explored.
Our aims
In this study, we aimed to conduct emotion and sentiment analysis on textual data retrieved from a public online platform serving as a cancer support group. Specifically, we were interested in the emotional experiences of those contemplating clinical trials. We approach their discussions as textual data and explore the verbal patterns representing emotional experiences expressed by users. We aimed to reveal the most prevalent basic emotions and sentiments, as well as complex emotions, in the posts of the patients and caregivers.
Second, we aimed to categorize the types of users and posts to understand better who uses this platform and what types of content are prevalent.
Methods and materials
Ethics statement
All posts included in the analysis are publicly available online data; therefore, ethical committee approval was not required for conducting this study. That said, we acknowledge the importance of the cautious use of public data. The use of social media data for research purposes is an emerging area, and the academic community is still developing strategies to ensure user confidentiality and consent [15, 16]. In view of these considerations, the analysis did not include names or identifiable information to ensure user anonymity.
Data collection
The data was retrieved from the subreddit ‘cancer’ with 45.9 K subscribers. Pushshift Reddit API Documentation [17] guidelines were used to create the search string and retrieve relevant posts. The keywords included in the search were the words that may refer to clinical trials: “clinical*trial”, “new*drug”, “new*treatment”, “experimental*trial”, “experimental*drug”, “experimental*treatment”. Together with the text body of the post, metadata was downloaded (epoch, username of the poster, ID of the post, web link). Epochs were converted into human-readable dates and times.
After downloading posts based on the keywords, the databases were compared, and the duplicates were deleted. Then, the first two authors assessed the remaining posts independently for final inclusion; posts that did not focus on clinical trials were excluded.
A total of 337 public posts between July 2011 to November 2021 were identified.
Statistical analysis
R studio v. 4.1.2 (2021-11-01) [18] was used for statistical analysis. Manual coding of each post was performed by two independent coders (C.M, C.F). This procedure consisted of labeling the type of users and posts with one of the predetermined labels. The preliminary labels (codes) were detected based on the 100 posts randomly selected by Google’s random generator. In cases of non-matching labels, the final label was chosen based on the discussion between the coders. Cohen’s kappa was calculated to determine inter-rater reliability. The values of Cohen’s kappa indicate different levels of agreement: 0 (no agreement), 0.01–0.20 (slight agreement), 0.21–0.40 (fair agreement), 0.41–0.60 (moderate agreement), 0.61–0.80 (substantial agreement), 0.81–1.00 (almost perfect agreement).
R package ‘Syuzhet’ and the NRC lexicon
The posts from patients and caregivers were analyzed with the R package ‘Syuzhet’ [19]; this package represents a sentiment extraction tool developed in the Stanford NLP group. It allows users to work on the sentiment data in their text files. Emotions and sentiments analysis was performed using the function ‘get_nrc_sentiment’, which considers eight basic emotions and two sentiments. This function uses NRC Emotion Lexicon [20, 21]. The lexicon is a list of words, each associated with corresponding emotions (anger, fear, anticipation, trust, surprise, sadness, joy, and disgust) and sentiments (positive and negative); this classification is in line with the work of Plutchik and his wheel of emotions [22]. The NRC lexicon has been increasingly applied in recent years for quantifying affect in online textual data [23–25].
With the use of ‘get_nrc_sentiment’ function, words of the data can be matched with identical words in the NRC lexicon and described by the emotions and sentiments they represent. Once the ‘get_nrc_sentiment’ function is applied to the text, a table is returned to the user, and later the returned table can be accessed and analyzed as any other data frame. This data frame consists of the split text arranged in rows, while the columns contain the corresponding sentiment and emotion association. The association’s presence can be determined by the number ‘0’ or ‘1’, where ‘0’ indicates that the word is not associated with the emotion or sentiment, whereas ‘1’ confirms the association. A word can be related to more than one emotion and may have a positive, negative, or polarity orientation. For example, words that embody anger, fear, disgust, and sadness are mostly associated with negative sentiment.
On the contrary, words expressing anticipation, joy, and trust are primarily associated with positive sentiment. Furthermore, surprise words can be either positive or negative, representing both polarities. Table 1 illustrates 10 example words from our dataset and how they are codified into the eight basic emotions in the lexicon.
Plutchik’s wheel of emotions
According to Plutchik [22], each of the basic emotions (out of 8) forms a complex emotion (dyad) when it is combined with one of the other 7 basic emotions. Plutchik visualized basic emotions on a wheel, allowing us to understand better how basic emotions are related to one another based on their spatial placement. Considering how far the two co-occurring emotions are, they form primary, secondary, and tertiary dyads or opposites.
Following this background, we extracted the frequencies of basic emotions to calculate the co-occurrence between emotions for creating the primary, secondary, tertiary, and opposite emotions’ dyads. Specifically, the co-occurrence in a sentence was represented by the presence of each specific emotion associated with another one. For example, joy and sadness are opposites, forming a complex emotion, ‘bitter sweetness’. Therefore, if in a sentence, the frequency of sadness is 2 and the frequency of joy is 3, the frequency of bitter sweetness was set to 2. Thus, the sum of each co-occurrence in a sentence was calculated as a representation of the frequency of a specific dyad.
To visualize the wheel, Python 3.10.5. was used [26]. We utilized the package ‘PyPlutchik’, a novel Python implementation designed to visualize Plutchik’s wheel of emotions [27]. For each wheel (basic emotions, primary dyads, secondary dyads, tertiary dyads, opposites) the emotion of the highest frequency was selected and set to 1. The scores of the rest of the emotions were calculated based on this ratio. As a result, emotions on the wheel range from 0 to 1, where 1 is the most frequent emotion.
Results
A total of 337 public posts between July 2011 to November 2021 were identified. Inter-rater reliability manual coding rated a perfect agreement (from .98 to .99) for both broad categories (see Table 2).
241 (129 from patients, 112 from caregivers) out of 337 comments were included in the final database since our aim was focused on patients and caregivers. Therefore, 96 comments were excluded from the analysis since the user type was unknown (N = 89) or referred to healthcare professionals (N = 7). See Table 2.
Sentiments and emotions analysis
Fig 1 shows the most prevalent emotions and sentiments in the posts of patients and caregivers when sharing advice, experience, or both regarding clinical trials. The total length of the words was 62061.
The negative polarity of emotions and sentiments is indicated in red, and the positive ones–in green. Both red and green were used for emotions that may be associated with both negative or positive emotions and sentiments. The count represents the total frequency of association between the word and one or more emotions and/or one of two polarities (negative and positive) present in the data body text.
Both positive (N = 2557, M = .68) and negative (N = 2154, M = .57) sentiments were high, with a slight prevalence of target words associated with the positive sentiment. Regarding the positive category, trust (N = 1470, M = .44) and anticipation (N = 1376, M = .37) were the two most prevalent emotions, followed by joy (N = 946, M = .25). Regarding the negative category, fear (N = 1702, M = .45) and sadness (N = 1494, M = .40) were found to be the highest, followed by anger (N = 1095, M = .29) and disgust (N = 925, M = .25).
On the contrary, users’ lowest affect expressed in the posts was surprise (N = 615, M = .16).
Tables 3 and 4 show examples of extracts from posts of users with the sentiments and emotions associated with each target word. Specifically, Table 3 presents an extract from a patient’s post sharing their experience. The rows represent sentences from their post, and the columns are created for emotions and sentiments. The numbers show how many times a specific emotion or sentiment was expressed in a given sentence. In Table 4, the columns represent emotions and sentiments. In contrast, the rows are examples (extracts) of different posts, post ID/user type, and the frequency of the specific emotion or sentiment (based on the column) detected in the example extract.
Plutchik’s wheel of emotions
The frequency of co-occurrence between pairs of emotions is reported in Table 5. The three most frequent primary dyads were remorse, contempt, and optimism, followed by submission, aggressiveness, love, alarm, and disappointment. Regarding the secondary dyads, the two most frequent were envy and despair, followed by hope, curiosity, guilt, cynism, pride, and unbelief. The most frequent tertiary dyad was outrage, followed by shame, anxiety, sentimentality, pessimism, delight, dominance, and morbidness. Finally, regarding the opposite dyads, we found frozenness to be the most frequent pair of emotions, followed by confusion, ambivalence, and bittersweetness. Refer to Fig 2 to view the wheel of emotions created by PyPlutchik Python package [27] based on our data.
The central wheel shows basic emotions, the wheel in the upper left corner shows primary dyads, the wheel in the lower left corner shows secondary dyads, the wheel in the upper right corner shows tertiary dyads, and the wheel in the lower right corner shows opposites.
Discussion
Online platforms provide a space for users to unload their tension and emotional discharge without taking responsibility for the effects this may have on others. In regard to emotional needs, social media platforms may satisfy the need for both emotional support (the communication between two or more users in which they mutually share and discuss their affective needs) and emotional expression (the freedom with which patients can disclose their experiences, regardless of if someone responds to them, without the concern that the shared experience will impact others around them) [28–30]. Addressing these needs is especially important, considering that some users, such as cancer patients and their caregivers, tackle emotionally charged challenges daily. Previous studies have explored posts on online cancer communities with methods such as topic analysis [11] and sentiment analysis [13, 14]. Our study differs from the previous works in two main aspects: (1) We narrowed our focus to the experiences of patients and caregivers discussing clinical trials. Due to their experimental, innovative, and at times risky nature, clinical trials may evoke particularly challenging emotions in users. Therefore, non-specific discussions in cancer communities may not fully reflect this topic’s nuance. (2) Given the complexity of these emotions, our analysis goes beyond the polarity of sentiments. In our analysis, we included the 8 basic emotions, following the theoretical framework of Plutchik and dyadic emotions. This way, we bridge the natural language processing algorithms with a psychological theoretical framework of emotions.
Overall, this study points out that users shared two main types of content in this online support group: personal experience and advice. Additionally, we found an almost equal distribution of patients and caregivers among posts with an identifiable user type. This, in our view, stresses that patients neither experience the clinical trial-related distress alone nor make decisions in isolation; in most cases, the caregivers are part of all the steps. Furthermore, caregivers may have an even higher burden than patients in trials [31], and elevated levels of commitment may impact the caregiver’s ability to provide care for the patient [32]. Therefore, we consider focusing on patients’ and caregivers’ emotions essential.
This study identified the most prevalent basic and complex emotions underlying users’ posts on the subreddit ‘cancer’. As represented in Plutchik’s wheel (Fig 2), the most prevalent basic emotions expressed in texts were fear, trust, and sadness. These emotions may correspond with the innovative nature of oncological clinical trials. Still, in the experimental phase, the drug’s side effects are not always predictable or obvious [33]. Therefore, trials represent the unknown and may elicit fear, concerns about trust towards the new drug or the physician responsible for the trial, and sadness resulting from the overall confusion. Previous works with patients in the trial or who are considering joining one inform us that patients often fear the side effects of clinical trials and the unknown surrounding these studies [34]. The fear that has been previously reported is about being treated as a “guinea pig” (being used for the trial and the career of the principal scientist, with no personal benefit) [35]. This fear was mirrored in our database as well; one of the caregivers shared:
“The clinical trial my brother is in has no efficacy because there were no trials. Basically, he’s a well-supported lab rat…”
This post not only represents the users’ concern but may also represent fear and trust issues towards the healthcare system and the sadness evoked by the confronted reality. Similar emotions may be elicited due to other concerns surrounding clinical trial participation. Among the barriers that hinder patients from participating are treatment-related concerns, fear of side effects, financial concerns, dislike for the experimental setting, and random assignment of patients to experimental and placebo groups [36].
Considering the complex emotions created by the dyads (Fig 2), we saw that among the most prevalent ones were (here we list those with a ratio above 0.8): remorse (composed of disgust and sadness); contempt (composed of anger and disgust); optimism (composed of anticipation and joy); envy (composed of anger and sadness); despair (composed of sadness and fear); outrage (composed of anger and surprise); and frozenness (composed of fear and anger). Most of these complex emotions (apart from optimism) are considered negative. This again emphasizes the struggles that patients and caregivers face and the emotional turmoil they go through. However, while studies focus on some psychological challenges faced by patients going through trials (e.g., lowered quality of life [37]), there are no studies investigating complex dyadic emotions, apart from optimism, in clinical trials. Some studies focus on optimism and clinical trials, specifically, unrealistic optimism about the outcomes of participation in a trial [38, 39].
We believe that the prevalence of dyadic emotions mirrors the complexity of clinical trial-related experiences. Clinical trials may have different meanings given to them by different people. While some relate it to being a “lab rat” (as in the post of a caregiver above), others may give it a completely different sense, such as the one provided by another user in our data:
“…I’ve had similar thoughts, and I’ve concluded that if I relapsed again, I would do a clinical trial still, not for myself, but because it would provide valuable data for future patients whether or not it works for me. Do with that what you will.”
Given that trials can be associated with altruistic narratives such as the one above (i.e., helping future patients), declining to participate in such research can elicit feelings of regret and guilt and the perception of personal moral failure [40, 41]. Moreover, this decision’s emotional burden may become even more complex due to the patient’s relationship with their oncologist. Many patients have a long relationship with their oncologist, and declining participation may elicit fear of harming this rapport or even cause worse treatment in the future [9]; such complex experiences may result in dyadic emotions (e.g., despair and frozenness).
Investigating complex emotions is crucial as it may reveal nuances that cannot be accessed by solely focusing on basic emotions. For example, an interesting observation we made in our results is about anger. Even though anger was not among the most frequently expressed basic emotions (being the fifth among eight basic emotions, Fig 1), within the most frequent dyadic couples (creating a complex emotion), anger is a recurring primary emotion. More specifically, anger is in the composition of the following: contempt, envy, outrage, and frozenness. This finding points to the fact that users experience anger about clinical trials in complex ways, often coupling it with other emotions. This may be due to the conceptual complexity of clinical trials; the diversity of the aspects that users consider when discussing them; suppression of direct expression of anger, and limiting it to be one of the experiences, among many others, etc.
We want to stress the importance of focusing on the emotional experiences of patients and their caregivers for further research. It is relevant not just for the overall well-being of these groups but also for gaining a deeper understanding of their behavioral patterns when deciding on their health. For example, it has been demonstrated among oncological patients that fear, anxiety, and worry interfere with medical decisions and behavior [42]. Understanding the mechanisms behind medical choices has critical implications on an individual patient level, as it can improve the ability of healthcare professionals to assist patients when confronted with a medical decision. Furthermore, considering the role of emotions in medical decision-making can affect the development of the field on a large scale. This is especially true for cancer clinical trials as the main gateway to developing new anti-cancer treatments. The most prevalent barriers to conducting a clinical trial are patients’ concerns about disease progression, experienced side effects, and dropout due to personal issues [43]. Therefore, it is crucial to consider the emotional experiences of patients and their caregivers, as they could interfere with their willingness to participate in or continue the clinical trial.
This study has some limitations. First, we did not compare patients’ and caregivers’ emotional experiences. Further research should consider this limit and assess emotions separately to explore their different perspectives.
Second, we needed access to the demographics and personal characteristics of Reddit users; therefore, differences based on these variables could not be considered. Moreover, within the methodological approach of this study, emotions cannot be linked to specific experiences about trials; in this respect, we can only speculate about possible explanations. This study aimed to reveal the prevalent emotions and sentiments rather than investigate their link to specific experiences. Nevertheless, this may present an interesting direction for future research.
Conclusions
In summary, although the barriers to participation in trials have been primarily investigated in the literature [3, 43], the nature of emotions standing behind patients’ and caregivers’ concerns has not been thoroughly explored. With this study, we address the question of the most prevalent basic and complex emotions among patients and caregivers discussing cancer clinical trials, emphasizing the importance and relevance of their diverse outlooks.
Finally, we would like to conclude this paper with a post from one of the users. The perspective portrayed in the post encapsulates the complexity of experiences in clinical trials and the complexity of emotions accompanying them:
“…The thing with trials is that there’s no guarantee that it’s going to work. It may extend your life, it may even shorten your life, or you may end up in the placebo group where you’re not actually getting the medicine on trial. Is it worth it? That’s really down to an individual’s feelings about it. Some people will continue to grasp at straws no matter how dire their chances may be. Some people would rather spend their remaining time with their families. Others want to contribute to the advancement of treatments so others after them may have better chances of survival.”
Acknowledgments
Mariam Chichua and Chiara Filipponi are PhD students within the European School of Molecular Medicine (SEMM).
References
- 1. Kelly C, Ghazi F, Caldwell K. Psychological distress of cancer and clinical trial participation: A review of the literature. Eur J Cancer Care (Engl). 2002;11:6–15. pmid:11966830
- 2. Ulrich CM, Knafl KA, Ratcliffe SJ, Richmond TS, Grady C, Miller-Davis C, et al. Developing a Model of the Benefits and Burdens of Research Participation in Cancer Clinical Trials. AJOB Prim Res. 2012;3:10–23. pmid:24748992
- 3. Mills EJ, Seely D, Rachlis B, Griffith L, Wu P, Wilson K, et al. Barriers to participation in clinical trials of cancer: A meta-analysis and systematic review of patient-reported factors. Lancet Oncology. 2006;7:141–8. pmid:16455478
- 4. Lewandowska A, Rudzki G, Lewandowski T, Rudzki S. The problems and needs of patients diagnosed with cancer and their caregivers. Int J Environ Res Public Health. 2021;18:1–19.
- 5. Northouse LL, Katapodi MC, Schafenacker AM, Weiss D. The Impact of Caregiving on the Psychological Well-Being of Family Caregivers and Cancer Patients. Semin Oncol Nurs. Elsevier Ltd; 2012;28:236–45. pmid:23107181
- 6. Dilworth S, Higgins I, Parker V, Kelly B, Turner J. Patient and health professional’s perceived barriers to the delivery of psychosocial care to adults with cancer: A systematic review. Psychooncology. 2014;23:601–12. pmid:24519814
- 7. Colineau N, Paris C. Talking about your health to strangers: Understanding the use of online social networks by patients. New Review of Hypermedia and Multimedia. 2010;16:141–60.
- 8. Pollak KI, Arnold RM, Jeffreys AS, Alexander SC, Olsen MK, Abernethy AP, et al. Oncologist communication about emotion during visits with patients with advanced cancer. Journal of Clinical Oncology. 2007;25:5748–52. pmid:18089870
- 9. Penner LA, Manning M, Eggly S, Albrecht TL. Prosocial behavior in cancer research: patient participation in cancer clinical trials. The Oxford handbook of prosocial behavior. 2015.
- 10. Rupert DJ, Moultrie RR, Read JG, Amoozegar JB, Bornkessel AS, O’Donoghue AC, et al. Perceived healthcare provider reactions to patient and caregiver use of online health communities. Patient Educ Couns. Elsevier Ireland Ltd; 2014;96:320–6. pmid:24923652
- 11. Tapi Nzali MD, Bringay S, Lavergne C, Mollevi C, Opitz T. What Patients Can Tell Us: Topic Analysis for Social Media on Breast Cancer. JMIR Med Inform [Internet]. 2017;5:e23. Available from: http://medinform.jmir.org/2017/3/e23/ pmid:28760725
- 12. Tan MY, Goh CE, Tan HH. Contemporary english pain descriptors as detected on social media using artificial intelligence and emotion analytics algorithms: Cross-sectional study. JMIR Form Res. 2021;5. pmid:34842554
- 13. Cabling ML, Turner JW, Hurtado-de-Mendoza A, Zhang Y, Jiang X, Drago F, et al. Sentiment Analysis of an Online Breast Cancer Support Group: Communicating about Tamoxifen. Health Commun [Internet]. Routledge; 2018;33:1158–65. Available from: pmid:28678549
- 14. Carrillo-de-Albornoz J, Vidal JR, Plaza L. Feature engineering for sentiment analysis in e-health forums. PLoS One. 2018;13:1–25. pmid:30496232
- 15. Proferes N, Walker S. Researcher Views and Practices around Informing, Getting Consent, and Sharing Research Outputs with Social Media Users When Using Their Public Data. 2020. Available from: https://hdl.handle.net/10125/64032
- 16. Ravn S, Barnwell A, Barbosa Neves B. What Is “Publicly Available Data”? Exploring Blurred Public–Private Boundaries and Ethical Practices Through a Case Study on Instagram. Journal of Empirical Research on Human Research Ethics [Internet]. 2020;15:40–5. Available from: http://journals.sagepub.com/doi/10.1177/1556264619850736 pmid:31132903
- 17. GitHub Inc. Pushshift Reddit API Documentation. 2022.
- 18. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria; 2021.
- 19. Jockers ML. Syuzhet: Extract Sentiment and Plot Arcs from Text. 2015.
- 20. Mohammad SM, Turney PD. Crowdsourcing a word-emotion association lexicon. Comput Intell. 2013;29:436–65.
- 21. Mohammad S, Turney P. Emotions Evoked by Common Words and Phrases: Using Mechanical Turk to Create an Emotion Lexicon, Proceedings of the {NAACL}-HLT 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text. Proceedings of the NAACL HLT 2010 Workshop on Computational Approaches to Analysis and Generation of Emotion in Text. 2018;14:26–34.
- 22. Plutchik R. The nature of emotions: Human emotions have deep evolutionary roots, a fact that may explain their complexity and provide tools for clinical practice. Am Sci. 2001;89:344–50.
- 23. Xavier T, Lambert J. Sentiment and emotion trends in nurses’ tweets about the COVID-19 pandemic. Journal of Nursing Scholarship. 2022;54:613–22. pmid:35343050
- 24. Ramezani M, Feizi-Derakhshi MR, Balafar MA. Knowledge Graph-Enabled Text-Based Automatic Personality Prediction. Comput Intell Neurosci. 2022;2022:3732351. pmid:35769270
- 25. Liu J, Gao L. Research on the characteristics and usefulness of user reviews of online mental health consultation services: A content analysis. Healthcare (Switzerland). 2021;9. pmid:34574885
- 26.
Van Rossum G, Drake FL. Python 3 Reference Manual. Scotts Valley, CA: CreateSpace; 2009.
- 27. Semeraro A, Vilella S, Ruffo G. PyPlutchik: Visualising and comparing emotion-annotated corpora. PLoS One. 2021;16:1–24. pmid:34469455
- 28. Smailhodzic E, Hooijsma W, Boonstra A, Langley DJ. Social media use in healthcare: A systematic review of effects on patients and on their relationship with healthcare professionals. BMC Health Serv Res. BMC Health Services Research; 2016;16:1–14.
- 29. Chiu YC, Hsieh YL. Communication online with fellow cancer patients: Writing to be remembered, gain strength, and find survivors. J Health Psychol. 2013;18:1572–81. pmid:23221492
- 30. Broom A. The eMale: Prostate cancer, masculinity and online support as a challenge to medical expertise. Journal of Sociology. 2005;41:87–104.
- 31. Rezash V, Reed J, Gedeon B, Parsons E, Siedlecki S, Daniels B, et al. Who Needs What? Perceptions of Patients and Caregivers in Oncology Phase 1 Trials. J Patient Exp. 2020;7:27–33. pmid:32128368
- 32. Stenberg U, Cvancarova M, Ekstedt M, Olsson M, Ruland C. Family Caregivers of Cancer Patients: Perceived Burden and Symptoms During the Early Phases of Cancer Treatment. Soc Work Health Care. 2014;53:289–309. pmid:24628120
- 33.
NCI NC institute. Informed Consent [Internet]. 2018 [cited 2022 Jun 28]. Available from: https://ctep.cancer.gov/protocoldevelopment/informed_consent.htm.
- 34. Quinn GP, Koskan A, Wells KJ, Gonzalez LE, Meade CD, Pratt Pozo CL, et al. Cancer patients’ fears related to clinical trial participation: A qualitative study. Journal of Cancer Education. 2012;27:257–62. pmid:22271582
- 35. Quinn GP, Bell BA, Bell MY, Caraway VD, Conforte D, Graci LB, et al. The guinea pig syndrome: Improving clinical trial participation among thoracic patients. Journal of Thoracic Oncology [Internet]. International Association for the Study of Lung Cancer; 2007;2:191–6. Available from: http://dx.doi.org/10.1097/JTO.0b013e318031cdb6
- 36. Unger JM, Hershman DL, Till C, Minasian LM, Osarogiagbon RU, Fleury ME, et al. “When Offered to Participate”: A Systematic Review and Meta-Analysis of Patient Agreement to Participate in Cancer Clinical Trials. J Natl Cancer Inst. 2021;113:244–57. pmid:33022716
- 37. Sidana S, Allmer C, Larson MC, Dueck A, Yost K, Warsame R, et al. Patient Experience in Clinical Trials: Quality of Life, Financial Burden, and Perception of Care in Patients With Multiple Myeloma or Lymphoma Enrolled on Clinical Trials Compared With Standard Care. JCO Oncol Pract. 0:OP.21.00789. pmid:35580285
- 38. Jansen LA, Appelbaum PS, Klein WMP, Weinstein ND, Cook W, Fogel JS, et al. Unrealistic optimism in early-phase oncology trials. IRB Ethics and Human Research. 2011;33:1–8. pmid:21314034
- 39. Jansen LA, Mahadevan D, Appelbaum PS, Klein WMP, Weinstein ND, Mori M, et al. Perceptions of control and unrealistic optimism in early-phase cancer trials. J Med Ethics. 2018;44:121–7. pmid:28774957
- 40. Stevens T, Ahmedzai SH. Why do breast cancer patients decline entry into randomised trials and how do they feel about their decision later: A prospective, longitudinal, in-depth interview study. Patient Educ Couns. 2004;52:341–8. pmid:14998605
- 41. Madsen SM, Holm S, Riis P. Participating in a cancer clinical trial? The balancing of options in the loneliness of autonomy: A grounded theory interview study. Acta Oncol (Madr). 2007;46:49–59.
- 42. Mazzocco K, Masiero M, Carriero MC, Pravettoni G. The role of emotions in cancer patients’ decision-making. Ecancermedicalscience. 2019;13. pmid:31123497
- 43. Kim J, Kim MG, Lim K-M. Participation in and withdrawal from cancer clinical trials: A survey of clinical research coordinators. Asia Pac J Oncol Nurs. 2022;9:197–201. pmid:35571627