Advertisement

2: Listening to the HysterSisters®: Online conversations about hysterectomy recovery

      Objectives

      Despite increased focus on patient-centered care, physicians remain pressured by multiple time constraints. As a result, less time may be spent discussing recovery expectations and other common but not life-threatening symptoms. However, these still significantly influence patient quality of life and experience. This creates an information gap that patients attempt to bridge, often through telephone calls. Increasingly, patients are turning to social media, such as online health communities to seek answers and reassurance. One such publicly accessible community, HysterSisters®, is “dedicated to medical and emotional issues surrounding the hysterectomy experience […], supporting women from diagnosis, to treatment, to recovery.” It offers a unique view into patients’ conversations surrounding hysterectomy and other gynecologic care. We performed a large-scale semantic analysis to identify common patient concerns during recovery from hysterectomy.

      Materials and Methods

      A “Hysterectomy Recovery” board on HysterSisters® was identified. A forum crawler was utilized to gather thread titles, all replies, and available metadata. Three-word clusters (“3-grams”) were tabulated for the thread titles and content of all replies. A signed chi-squared residual test was performed to compare the expected frequencies of the tabulated 3-grams in the titles as compared to the posts.

      Results

      We catalogued 851,443 individual posts in 122,372 unique threads and identified 73,666 unique 3-grams meeting inclusion criteria for statistical analysis. Three-grams with “hysterectomy” as the first word were selected for analysis, of which 1,243 were identified. Among the top 50 3-grams, 60% were related to symptoms (e.g. “can’t sleep”), 20% to patient question stems (e.g. “how long”), 10% to concerns about persistent pain, bleeding and fatigue (e.g. “still bleed”), and 8% to other issues. Abdominal swelling ranked just below back pain and received its own moniker in the community, the “Swelly Belly.” Additional notable symptom-related 3-grams referenced “can’t sleep,” “hot flash,” “feel blue,” “mood swing,” and “yeast infection.” Thirty-four percent (17/50) of the top 50 3-grams dealt directly with pain complaints, including most commonly pain with bowel movements, in the lower back, bladder and breast, as well as pain after activity and localized pains (“left” or “right”). Among the top 50, 3-grams were also “feel great” and “feel worse.” Patient question stems included “how long,” “anyone else,” and “when can” among others. Other issues included “steri strip,” “pain med,” “vaginal cuff,” and “back [to] work.” The most common 3-grams in the post replies included “you feel better,” “feel better soon,” and “call your doctor.” Patient-to-patient support is present as well with respondents encouraging the thread creator to “take stool soften” or “take pain med” and reassuring that it “does get better.”

      Conclusion

      Although these issues may be a “normal” part of the recovery process, patients still frequently seek to bridge knowledge gaps. Important areas for counseling are revealed and can help physicians provide truly patient-centered care. This large-scale analysis of social media conversation data can offer unique insights into the patient experience.