Evidano home
Commentary on News

Qualitative Social Media Analysis: Cancer Communities

Evidano8 min read

Researchers studying patient advocacy, healthcare communication, and precision oncology need to understand more than message volume. They need to identify who participates, which themes emerge, how information moves between stakeholder groups, and how online conversations change over time. Qualitative social media analysis provides that context by connecting large-scale patterns with close reading of conversations. A 2026 PLoS One study demonstrates the value of this approach through a digital ethnography of cancer communities on Twitter, using more than 8.6 million publications collected from activity spanning 2010 to 2022. This article explains the study's findings and shows how AI-enabled qualitative research can help teams organize large document collections, compare stakeholder segments, and examine representative conversations without treating automated analysis as a substitute for research judgment.

Key Takeaways

According to the 2026 PLoS One study, qualitative social media analysis can explain how cancer advocates, oncologists, patients, and media organizations exchange knowledge, shape research concerns, and influence discussions about precision oncology.

The study combined semantic analysis, user classification, retweet-network analysis, and close qualitative reading to connect broad community structures with detailed conversations about biomarkers such as ROS1, ALK, EGFR, and HER2.

  • Between 2010 and 2022, the study assembled a dataset containing 8, 614, 284 publications, including 4, 389, 619 tweets and 3, 031, 281 retweets.
  • In 2023, the researchers selected 834 highly active users whose posts represented more than 80% of tweets containing the study's cancer-related hashtags, although those users represented less than 3% of 28, 524 users.
  • The study identified 116 initial topic clusters, consolidated them into 49 topic groups, and organized them into 13 overarching categories.
  • Precision oncology represented 2.4% of the corpus, while cancer-related topics represented 50% of the corpus after excluding outliers.
  • The researchers wrote that the qualitative approach showed Twitter could become a discussion space where participants developed 'concrete answers to patients’ questions' and policy proposals about care organization and clinical-trial regulation.

How Qualitative Social Media Analysis Worked in the Cancer Study

The 2026 PLoS One study used a mixed-method design that combined computational breadth with qualitative interpretation. The researchers began with approximately 30 cancer-related hashtags, then expanded the analysis to the complete tweet histories of 834 highly active accounts collected in 2023.

The researchers manually classified the 834 accounts into eight stakeholder categories: patients, advocates, oncologists, health professionals, health centers, media, nonprofit organizations, and industries. Four researchers reviewed account handles and biographies, and they discussed ambiguous cases until reaching consensus.

For semantic analysis, the study used BERTopic with transformer-based embeddings, UMAP dimensionality reduction, and HDBSCAN clustering. The researchers then reviewed the 50 most representative tweets for each topic to validate coherence and assign meaningful labels.

The study also used retweet networks to examine interaction patterns between stakeholder groups. The researchers found that advocates and oncologists both showed homophilic behavior, meaning they tended to amplify content from their own groups, although the pattern was more pronounced among oncologists.

The researchers paired those large-scale findings with a qualitative review of 7, 670 tweets organized into approximately 3, 000 conversations involving advocate Janet Freeman-Daily. The close reading examined how ROS1-related discussions connected patient experiences, oncology expertise, clinical trials, drug access, and advocacy-group formation.

Janet Freeman-Daily described Twitter as 'the one place where all stakeholders […] could meet and exchange ideas on something that had life-and-death implications for so many.' The quotation illustrates why qualitative interpretation matters: network structure can show connections, but conversation analysis can clarify what participants were trying to accomplish.

Cancer Social Media Analysis Findings at a Glance

Date or periodMetricValueResearch implication
2010 to 2022Publication history analyzed8, 614, 284 publicationsA decade-long corpus supports longitudinal analysis of changing cancer conversations.
2023Highly active accounts selected834 of 28, 524 usersA small, active core generated more than 80% of hashtag-related tweets.
2010 to 2022Topic clusters116 initial clusters, 49 groups, and 13 overarching categoriesHierarchical coding can preserve detail while producing usable themes.
2010 to 2022Precision oncology content2.4% of tweetsA focused biomarker analysis can isolate a smaller conversation within a broad health corpus.
2010 to 2022Cancer-related content50% of the corpus excluding outliersA broad collection can include both disease-specific and everyday community interaction.
July 2013 to December 2021Lung cancer social-media chatsMore than 150 chatsRepeated moderated discussions created a recurring forum for cross-stakeholder exchange.

Implications for Patient Advocacy and Health Research Teams

Qualitative social media analysis helps health researchers distinguish information diffusion from community building. The 2026 PLoS One study found that 25% of labeled tweets concerned mundane topics, showing that everyday interaction was part of the relationship infrastructure supporting information exchange.

Patient advocates played a bridging role in the study because advocates engaged with a wider range of stakeholder groups than patients and amplified information from medical sources. The researchers reported that advocates posted relatively little about HER2 but became highly active in ROS1 discussions between 2016 and 2018.

The ROS1 case demonstrates why longitudinal analysis matters for policy and patient-engagement research. The study linked increased ROS1 advocacy activity with the creation of The ROS1ders in 2015 and the United States Food and Drug Administration's approval of crizotinib for ROS1-positive metastatic non-small-cell lung cancer on March 11, 2016.

The study also identified important sampling limits. The 834 selected accounts represented less than 3% of users but more than 80% of hashtag-related tweets, so the findings describe the highly engaged core more strongly than passive or peripheral participants.

The authors reported that the corpus was predominantly Anglo-American and centered on United States regulatory and oncology settings. Researchers applying the findings to other countries, languages, platforms, or healthcare systems should treat transferability as an empirical question rather than an assumption.

For health and policy teams, the practical lesson is to combine frequency and network measures with close reading. Counts can identify where activity concentrates, while qualitative coding can reveal unmet needs, access barriers, stakeholder disagreements, and the meanings participants attach to biomedical innovation.

How Evidano Helps With AI-Enabled Qualitative Social Media Analysis

Problem: Large social-media corpora are difficult to review manually

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze social-media content, documents, interviews, and survey responses.

Researchers can upload collected text and use thematic, content, frequency, and cross-segment analyses to identify recurring concerns, stakeholder differences, and changes across time periods. The workflow supports the same broad-to-specific logic used in the 2026 PLoS One study, while keeping human researchers responsible for interpretation and validation.

Teams can explore the platform's relevant capabilities through Evidano's features.

Problem: Stakeholder groups need comparable qualitative profiles

Evidano can support cross-segment analysis by comparing themes across patient, advocate, clinician, media, or organizational segments represented in a dataset.

Researchers can use hierarchical codes and subcodes to organize broad topics such as precision oncology into narrower themes such as biomarker education, trial access, treatment side effects, and policy concerns.

Problem: Researchers need answers grounded in their source material

Evidano's AI chat lets research teams ask questions across uploaded documents and analyses, helping them locate supporting passages and compare interpretations across a corpus.

A researcher could ask which stakeholder groups discuss clinical-trial access, how ROS1 conversations differ from HER2 conversations, or which themes appear before and after a defined event. Researchers should still audit source passages, sampling decisions, and coding quality before reporting findings.

Problem: Sensitive health-related data requires careful handling

Evidano encrypts data and states that customer data is never used to train third-party models. Researchers working with patient narratives should also apply consent, anonymization, access-control, and platform-policy requirements appropriate to their study.

FAQ: Qualitative Social Media Analysis in Cancer Research

What is qualitative social media analysis?

Qualitative social media analysis is the systematic interpretation of online posts and interactions to understand meanings, experiences, relationships, and social processes.

The 2026 PLoS One study combined qualitative reading with topic modeling, user classification, and retweet-network analysis to explain how cancer stakeholders exchanged knowledge.

How did the cancer study analyze Twitter data?

The researchers analyzed a Twitter corpus spanning 2010 to 2022, classified 834 highly active accounts, modeled topics, examined retweet networks, and closely read selected conversations.

The researchers used BERTopic, UMAP, and HDBSCAN for computational topic discovery, then reviewed representative tweets qualitatively.

What did the study find about patient advocates?

The study found that patient advocates connected patients, oncologists, media, and other organizations by engaging with a broader range of sources than many other stakeholder groups.

The researchers described advocates as potential bridges between medical communities and patients, while noting that the network evidence was descriptive rather than proof of causation.

Why is qualitative analysis necessary when researchers already have topic models?

Qualitative analysis is necessary because topic models can identify clusters but cannot independently explain the context, intent, power relationships, or practical consequences of individual exchanges.

The 2026 PLoS One study used computational methods to identify areas of interest and close reading to interpret conversations about treatment access, clinical trials, and patient advocacy.

Can AI replace human researchers in health-related qualitative research?

AI can accelerate organization, coding, comparison, and retrieval, but AI should not replace human judgment in health-related qualitative research.

Researchers must review representative excerpts, document sampling limits, protect personal data, and distinguish descriptive associations from causal conclusions. This content is research-focused and is not medical advice or a diagnostic tool.

Conclusion and Next Steps

The 2026 PLoS One study shows that qualitative social media analysis can connect millions of online publications with the lived experience, advocacy work, and policy concerns behind them. The study's decade-long corpus, 834-account stakeholder classification, 13 topic categories, and close reading of ROS1 conversations demonstrate why mixed methods are valuable for health and patient-engagement research.

AI-enabled qualitative research can make this workflow more manageable by helping teams organize large text collections, compare stakeholder segments, identify recurring themes, and retrieve evidence for human review. Evidano supports that workflow with document analysis, cross-segment analysis, hierarchical coding, and AI chat over research materials.

Start turning complex social-media and qualitative datasets into structured, reviewable insight with Try Evidano for free.

Topics

  • qualitative social media analysis
  • AI qualitative analysis
  • digital ethnography cancer communities
  • patient advocacy research
  • precision oncology social media analysis

Keep reading

Browse all articles