Using Health Related Social MEDIA to Understand Experiences of Adults with LUNG Cancer in the Era of Immune-Oncology and Targeted Therapies
Author(s)
Booth A1, Manson S2, Halhol S1, Merinopoulou E1, Raluy-Callado M1, Hareendran A1, Knoll S2
1Evidera, London, UK, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA
Background: Non-small cell lung cancer (NSCLC) outcomes have evolved dramatically with the approval of immune-oncology (IO) and targeted therapies (TT). Commonly used patient reported outcome (PRO) tools were developed prior to such advancements. This study aimed to understand experiences of NSCLC patients receiving chemotherapy, IO or TT using disease-specific social media to inform future PRO development. Methods: Publicly available patient/caregiver (users) posts (2010-2019) were extracted from five lung cancer/NSCLC-specific websites. Content was deidentified to protect patient privacy. Users were stratified by metastatic and adjuvant-eligible subgroups and treatment received using lexical terms. Machine learning was used to reduce noise in the data. Symptom mentions were captured using natural language processing. Qualitative data analysis (QDA) was conducted on random samples of posts mentioning pain-related, fatigue-related, respiratory or infection-related symptoms. Results: Overall, 7,466 users (72,310 posts) were included (metastatic: 2,632 [53,784 posts]; adjuvant: 636 [5,298 posts]). Among all metastatic users, pain/discomfort and fatigue (including weakness) were the most commonly mentioned symptoms (49.7% and 39.6%, respectively). In the metastatic QDA (258 posts from 134 users), most frequently reported impacts included physical impairments, impacts on sleep, and eating. Among all adjuvant users, pain/discomfort and respiratory symptoms were the most common (44.8% and 23.9%, respectively). The adjuvant QDA (54 posts from 92 users) symptom impacts were also mostly on physical functioning. Positive treatment impacts (such as the ability to do routine activities) were also mentioned by both metastatic and adjuvant users. Discussion: This study incorporated patient experiences and uncovered symptoms most commonly discussed by patients. Online community users most frequently mentioned pain and fatigue which may need consideration as patient relevant outcomes/endpoints in addition to positive impacts on day-to-day life. Findings could be explored further and validated in future research with patients to ensure outcomes/endpoints are reflective of the current treatment experience.
Conference/Value in Health Info
2020-11, ISPOR Europe 2020, Milan, Italy
Value in Health, Volume 23, Issue S2 (December 2020)
Code
PCN302
Topic
Methodological & Statistical Research, Patient-Centered Research
Topic Subcategory
Artificial Intelligence, Machine Learning, Predictive Analytics, Patient-reported Outcomes & Quality of Life Outcomes, PRO & Related Methods, Stated Preference & Patient Satisfaction
Disease
Oncology