Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Nov 10;10(11):e37945.
doi: 10.2196/37945.

The Real-World Experiences of Persons With Multiple Sclerosis During the First COVID-19 Lockdown: Application of Natural Language Processing

Deborah Chiavi #  1 , Christina Haag #  1   2 , Andrew Chan  3 , Christian Philipp Kamm  3   4 , Chloé Sieber  1   2 , Mina Stanikić  1   2 , Stephanie Rodgers  1 , Caroline Pot  5 , Jürg Kesselring  6 , Anke Salmen  3 , Irene Rapold  1 , Pasquale Calabrese  7 , Zina-Mary Manjaly  8   9 , Claudio Gobbi  10   11 , Chiara Zecca  10   11 , Sebastian Walther  12 , Katharina Stegmayer  12 , Robert Hoepner  3 , Milo Puhan  1 , Viktor von Wyl  1   2
Affiliations

The Real-World Experiences of Persons With Multiple Sclerosis During the First COVID-19 Lockdown: Application of Natural Language Processing

Deborah Chiavi et al. JMIR Med Inform. .

Abstract

Background: The 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.

Objective: We 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.

Methods: We 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.

Results: A 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.

Conclusions: This 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.

Keywords: COVID-19; clinical informatics; health data; linguistic inquiry; medical informatics; multiple sclerosis; natural language processing; nervous system disease; nervous system disorder; patient data; sentiment analysis; textual data; topic modeling.

PubMed Disclaimer

Conflict of interest statement

Conflicts of Interest: CPK has received honoraria for lectures as well as research support from Biogen, Novartis, Almirall, Bayer Schweiz AG, Teva, Merck, Sanofi Genzyme, Roche, Eli Lilly, Celgene, and the Swiss Multiple Sclerosis (MS) Society (SMSG). AS has received speaker honoraria and/or travel compensation for activities with Almirall Hermal GmbH, Biogen, Merck, Novartis, Roche, and Sanofi Genzyme, and research support from the Swiss MS Society, none related to this work. The employer Department of Neurology, Regional Hospital Lugano [EOC], Lugano, Switzerland received financial support for CZ and CG's speaking and educational, research, or travel grants from Abbvie, Almirall, Biogen Idec, Celgene, Sanofi, Merck, Novartis, Teva Pharma, and Roche. AC has received speaker/board honoraria from Actelion (Janssen/J&J), Almirall, Bayer, Biogen, Celgene (BMS), Genzyme, Merck KGaA (Darmstadt, Germany), Novartis, Roche, and Teva, all for hospital research funds. He received research support from Biogen, Genzyme, UCB, the European Union, and the Swiss National Foundation. He serves as associate editor of the European Journal of Neurology, is on the editorial board for Clinical and Translational Neuroscience, and serves as topic editor for the Journal of International Medical Research. RH has received honoraria from Janssen, Lundbeck, Mepha, and Neurolite. SW has received honoraria from Janssen, Lundbeck, Mepha, Neurolite, and Sunovion. MS reports employment by Roche from February 2019 to February 2020. KS has received honoraria from from Janssen, Lundbeck and Mepha.

Figures

Figure 1
Figure 1
Flow diagram displaying the assessment procedure and subsequent selection procedure for online participants. Only online participants who described the experienced impact of COVID-19 on their personal life with at least 10 words were included in the text analysis.
Figure 2
Figure 2
Survey responses included in this study. (A) Histogram depicting the text entries of different word lengths on the self-reported daily-life impact of COVID-19 (n=885). The number of words per text entry are plotted along the y-axis. (B) Amount of completed surveys across time (April 8, 2020, to August 27, 2020). Overall, 86.9% (555/639) of the responses were collected during the first lockdown (ie, before April 27, 2020). The number of completed surveys is displayed on the y-axis. Time (ie, days) is plotted along the x-axis.
Figure 3
Figure 3
Most frequent keywords across free-text descriptions on participants’ perceived impact of COVID-19 on their personal life. Only text entries with at least 10 words in total were considered (n=639). “Stop words” (eg, “and” and “the”) were removed prior to the analysis.
Figure 4
Figure 4
Word cloud visualizing the most frequent keywords related to the impact of COVID-19 on participants’ personal lives across the complete study sample. Word size reflects the relative frequency of a specific word in comparison to the total number of analyzed words. Only text entries with at least 10 words in total were considered (n=639).
See this image and copyright information in PMC

References

    1. Cammel SA, De Vos MS, van Soest D, Hettne KM, Boer F, Steyerberg EW, Boosman H. How to automatically turn patient experience free-text responses into actionable insights: a natural language programming (NLP) approach. BMC Med Inform Decis Mak. 2020 May 27;20(1):97. doi: 10.1186/s12911-020-1104-5. https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-0... 10.1186/s12911-020-1104-5 - DOI - DOI - PMC - PubMed
    1. Dreisbach C, Koleck TA, Bourne PE, Bakken S. A systematic review of natural language processing and text mining of symptoms from electronic patient-authored text data. Int J Med Inform. 2019 May;125:37–46. doi: 10.1016/j.ijmedinf.2019.02.008. https://europepmc.org/abstract/MED/30914179 S1386-5056(18)31378-9 - DOI - PMC - PubMed
    1. Koleck TA, Dreisbach C, Bourne PE, Bakken S. Natural language processing of symptoms documented in free-text narratives of electronic health records: a systematic review. J Am Med Inform Assoc. 2019 Apr 01;26(4):364–379. doi: 10.1093/jamia/ocy173. https://europepmc.org/abstract/MED/30726935 5307912 - DOI - PMC - PubMed
    1. Mascio A, Kraljevic Z, Bean D, Dobson R, Stewart R, Bendayan R, Roberts A. Comparative Analysis of Text Classification Approaches in Electronic Health Records. arXiv. 2005. [2021-11-18]. http://arxiv.org/abs/2005.06624 .
    1. Calvo RA, Milne DN, Hussain MS, Christensen H. Natural language processing in mental health applications using non-clinical texts. Nat. Lang. Eng. 2017 Jan 30;23(5):649–685. doi: 10.1017/S1351324916000383. - DOI

LinkOut - more resources