I previously posted about
this study; however, the Full Text has now been published. This Forum is listed first in Fig. 1 (Online Support Groups). Table 2 has the primary treatment as determined for approx.18,500 patients across sites: RP 63%, RT 32%, and AS 5%.
Can online support groups address psychological morbidity of cancer patients? An artificial intelligence based investigation of prostate cancer trajectories (2020, Full Text)
And this is an editorial in
Nature Reviews Urology about
the study:
You’ve got a friend online (2020)
"The first study to empirically assess psychological morbidity and emotions expressed by men newly diagnosed with prostate cancer who use online cancer support groups using artificial intelligence has been published in PLoS One. Participation in these groups decreased psychological stress and long-term engagement improved emotional well-being. Thus, men with newly diagnosed prostate cancer might benefit from joining an online support group.
In this study, the researchers used artificial intelligence to extract and analyse activities of men with prostate cancer using online support groups in order to assess the reasons that these men joined, the emotions they expressed and the variation in deep emotions from diagnosis through to recovery, which could indicate psychological morbidity. To extract this data, the investigators extended the Patient-Reported Information Multidimensional Exploration (PRIME) framework. PRIME automatically detected relevant data (emotions expressed and mentions of adverse effects) from 277,805 conversations that were conducted by 18,496 patients in ten globally available, high-volume online support groups. This data collection enabled the creation of an emotion and adverse-effect profile for each patient who took part in the groups. The reasons for joining the online support group were assessed using natural language processing and machine-learning-based automatic topic extraction.
Four distinct reasons for joining a support group were identified: diagnosis; treatment; adverse effects; and cancer recurrence. Information on treatment was the most common reason for joining a group (61.72%). The most frequently expressed emotions in the first post were open and interested (which were classed as positive) and afraid and hurt (which were classed as negative). When expressed emotions were analysed in the context of the reasons for joining, ‘interested’ was associated with adverse effects, ‘positive’ was expressed by men whose reason for joining was diagnosis, and all reasons except adverse effects were associated with expressing ‘afraid’ as an emotion. Joining for adverse effects was associated with increased emotions related to ‘hurt’ and ‘sad’.
Evaluation of emotional fluctuations showed that negative emotions reduced in intensity after 12 months. For all reasons for joining except cancer recurrence, negative emotions were significantly reduced at 12 months after joining compared with at 1 month. Joining after treatment was associated with higher negative emotion intensity than joining before treatment, even at 12 months despite a reduction in negative emotions for both groups.
These results suggest that clinicians should inform newly diagnosed patients about online support groups as they can improve the psychological well-being of men with prostate cancer."
[Emphasis mine]
Djin