Search

BackgroundThe increasing availability of “real-world” data in the form of written text holds promise for deepening our understanding of societal and health-related challenges. Textual data constitute a rich source of information, allowing the capture of lived experiences through a broad range of different sources of information (eg, content and emotional tone). Interviews are the “gold standard” for gaining qualitative insights into individual experiences and perspectives. However, conducting interviews on a large scale is not always feasible, and standardized quantitative assessment suitable for large-scale application may miss important information. Surveys that include open-text assessments can combine the advantages of both methods and are well suited for the application of natural language processing (NLP) methods. While innovations in NLP have made large-scale text analysis more accessible, the analysis of real-world textual data is still complex and requires several consecutive steps. ObjectiveWe developed and subsequently examined the utility and scientific value of an NLP pipeline for extracting real-world experiences from textual data to provide guidance for applied researchers. MethodsWe applied the NLP pipeline to large-scale textual data collected by the Swiss Multiple Sclerosis (MS) registry. Such textual data constitute an ideal use case for the study of real-world text data. Specifically, we examined 639 text reports on the experienced impact of the first COVID-19 lockdown from the perspectives of persons with MS. The pipeline has been implemented in Python and complemented by analyses of the “Linguistic Inquiry and Word Count” software. It consists of the following 5 interconnected analysis steps: (1) text preprocessing; (2) sentiment analysis; (3) descriptive text analysis; (4) unsupervised learning–topic modeling; and (5) results interpretation and validation. ResultsA topic modeling analysis identified the following 4 distinct groups based on the topics participants were mainly concerned with: “contacts/communication;” “social environment;” “work;” and “errands/daily routines.” Notably, the sentiment analysis revealed that the “contacts/communication” group was characterized by a pronounced negative emotional tone underlying the text reports. This observed heterogeneity in emotional tonality underlying the reported experiences of the first COVID-19–related lockdown is likely to reflect differences in emotional burden, individual circumstances, and ways of coping with the pandemic, which is in line with previous research on this matter. ConclusionsThis study illustrates the timely and efficient applicability of an NLP pipeline and thereby serves as a precedent for applied researchers. Our study thereby contributes to both the dissemination of NLP techniques in applied health sciences and the identification of previously unknown experiences and burdens of persons with MS during the pandemic, which may be relevant for future treatment.

BackgroundElectronic health diaries hold promise in complementing standardized surveys in prospective health studies but are fraught with numerous methodological challenges. ObjectiveThe study aimed to investigate participant characteristics and other factors associated with response to an electronic health diary campaign in persons with multiple sclerosis, identify recurrent topics in free-text diary entries, and assess the added value of structured diary entries with regard to current symptoms and medication intake when compared with survey-collected information. MethodsData were collected by the Swiss Multiple Sclerosis Registry during a nested electronic health diary campaign and during a regular semiannual Swiss Multiple Sclerosis Registry follow-up survey serving as comparator. The characteristics of campaign participants were descriptively compared with those of nonparticipants. Diary content was analyzed using the Linguistic Inquiry and Word Count 2015 software (Pennebaker Conglomerates, Inc) and descriptive keyword analyses. The similarities between structured diary data and follow-up survey data on health-related quality of life, symptoms, and medication intake were examined using the Jaccard index. ResultsCampaign participants (n=134; diary entries: n=815) were more often women, were not working full time, did not have a higher education degree, had a more advanced gait impairment, and were on average 5 years older (median age 52.5, IQR 43.25-59.75 years) than eligible nonparticipants (median age 47, IQR 38-55 years; n=524). Diary free-text entries (n=632; participants: n=100) most often contained references to the following standard Linguistic Inquiry and Word Count word categories: negative emotion (193/632, 30.5%), body parts or body functioning (191/632, 30.2%), health (94/632, 14.9%), or work (67/632, 10.6%). Analogously, the most frequently mentioned keywords (diary entries: n=526; participants: n=93) were “good,” “day,” and “work.” Similarities between diary data and follow-up survey data, collected 14 months apart (median), were high for health-related quality of life and stable for slow-changing symptoms such as fatigue or gait disorder. Similarities were also comparatively high for drugs requiring a regular application, including interferon beta-1a (Avonex) and glatiramer acetate (Copaxone), and for modern oral therapies such as fingolimod (Gilenya) and teriflunomide (Aubagio). ConclusionsDiary campaign participation seemed dependent on time availability and symptom burden and was enhanced by reminder emails. Electronic health diaries are a meaningful complement to regular structured surveys and can provide more detailed information regarding medication use and symptoms. However, they should ideally be embedded into promotional activities or tied to concrete research study tasks to enhance regular and long-term participation.