Patient-Reported Outcomes of Opioid-Induced Constipation as Identified Through Social Media

Abstract

The aim of this exploratory project was to better understand the patient-reported impact of opioid-induced constipation (OIC) as discussed in social media. In this project, we applied text mining and machine-learning techniques to interrogate social media data. A web crawler was seeded with some initial URLs of websites likely to be relevant to the content we were seeking; further websites were then identified by the crawler. A machine-learning application was developed to extract from the web-pages anonymised patient-reported concepts relating to the burden of OIC on patients. Of the 42,000 web-pages retrieved, 122 described the patient-reported impact of OIC. There were 40 posts that spoke about the impact of OIC on dietary habits or the need for laxatives (with an additional 6 posts speaking about the need for enemas or manual excavations). There were 36 posts describing the emotional impact (including worry) associated with OIC, with 23 posts describing how OIC had necessitated a change in opioid use. The impact of OIC on general health and everyday functioning (including work) were other topics that patients discussed (9 and 8 posts, respectively). Social media posts can offer an important source of data on patient experiences of disease. In this exploratory study, these posts have indicated that OIC impacts a number of different aspects of people’s lives, including general health, daily functioning, and emotional well-being. OIC also impacts behaviour (dietary change, change in opioid use, and need for procedures) as patients attempt to reduce its burden. Findings from this study could be used as an input to the development of a conceptual model, or to supplement other qualitative research on the patient-reported impact of OIC.