COVID-19 Vaccine Hesitancy and Acceptance Among Individuals With Cancer, Autoimmune Diseases, or Other Serious Comorbid Conditions: Cross-sectional, Internet-Based Survey.

BACKGROUND: Individuals with comorbid conditions have been disproportionately affected by COVID-19. Since regulatory trials of COVID-19 vaccines excluded those with immunocompromising conditions, few patients with cancer and autoimmune diseases were enrolled. With limited vaccine safety data available, vulnerable populations may have conflicted vaccine attitudes. OBJECTIVE: We assessed the prevalence and independent predictors of COVID-19 vaccine hesitancy and acceptance among individuals with serious comorbidities and assessed self-reported side effects among those who had been vaccinated. METHODS: We conducted a cross-sectional, 55-item, online survey, fielded January 15, 2021 through February 22, 2021, among a random sample of members of Inspire, an online health community of over 2.2 million individuals with comorbid conditions. Multivariable regression analysis was utilized to determine factors independently associated with vaccine hesitancy and acceptance. RESULTS: Of the 996,500 members of the Inspire health community invited to participate, responses were received from 21,943 individuals (2.2%). Respondents resided in 123 countries (United States: 16,277/21,943, 74.2%), had a median age range of 56-65 years, were highly educated (college or postgraduate degree: 10,198/17,298, 58.9%), and had diverse political leanings. All respondents self-reported at least one comorbidity: cancer, 27.3% (5459/19,980); autoimmune diseases, 23.2% (4946/21,294); chronic lung diseases: 35.4% (7544/21,294). COVID-19 vaccine hesitancy was identified in 18.6% (3960/21,294), with 10.3% (2190/21,294) declaring that they would not, 3.5% (742/21,294) stating that they probably would not, and 4.8% (1028/21,294) not sure whether they would agree to be vaccinated. Hesitancy was expressed by the following patients: cancer, 13.4% (731/5459); autoimmune diseases, 19.4% (962/4947); chronic lung diseases: 17.8% (1344/7544). Positive predictors of vaccine acceptance included routine influenza vaccination (odds ratio [OR] 1.53), trust in responsible vaccine development (OR 14.04), residing in the United States (OR 1.31), and never smoked (OR 1.06). Hesitancy increased with a history of prior COVID-19 (OR 0.86), conservative political leaning (OR 0.93), younger age (OR 0.83), and lower education level (OR 0.90). One-quarter (5501/21,294, 25.8%) had received at least one COVID-19 vaccine injection, and 6.5% (1390/21,294) completed a 2-dose series. Following the first injection, 69.0% (3796/5501) self-reported local reactions, and 40.0% (2200/5501) self-reported systemic reactions, which increased following the second injection to 77.0% (1070/1390) and 67.0% (931/1390), respectively. CONCLUSIONS: In this survey of individuals with serious comorbid conditions, significant vaccine hesitancy remained. Assumptions that the most vulnerable would automatically accept COVID-19 vaccination are erroneous and thus call for health care team members to initiate discussions focusing on the impact of the vaccine on an individual's underlying condition. Early self-reported side effect experiences among those who had already been vaccinated, as expressed by our population, should be reassuring and might be utilized to alleviate vaccine fears. Health care-related social media forums that rapidly disseminate accurate information about the COVID-19 vaccine may play an important role.

Early Detection of Adverse Drug Reactions in Social Health Networks: A Natural Language Processing Pipeline for Signal Detection.

BACKGROUND: Adverse drug reactions (ADRs) occur in nearly all patients on chemotherapy, causing morbidity and therapy disruptions. Detection of such ADRs is limited in clinical trials, which are underpowered to detect rare events. Early recognition of ADRs in the postmarketing phase could substantially reduce morbidity and decrease societal costs. Internet community health forums provide a mechanism for individuals to discuss real-time health concerns and can enable computational detection of ADRs. OBJECTIVE: The goal of this study is to identify cutaneous ADR signals in social health networks and compare the frequency and timing of these ADRs to clinical reports in the literature. METHODS: We present a natural language processing-based, ADR signal-generation pipeline based on patient posts on Internet social health networks. We identified user posts from the Inspire health forums related to two chemotherapy classes: erlotinib, an epidermal growth factor receptor inhibitor, and nivolumab and pembrolizumab, immune checkpoint inhibitors. We extracted mentions of ADRs from unstructured content of patient posts. We then performed population-level association analyses and time-to-detection analyses. RESULTS: Our system detected cutaneous ADRs from patient reports with high precision (0.90) and at frequencies comparable to those documented in the literature but an average of 7 months ahead of their literature reporting. Known ADRs were associated with higher proportional reporting ratios compared to negative controls, demonstrating the robustness of our analyses. Our named entity recognition system achieved a 0.738 microaveraged F-measure in detecting ADR entities, not limited to cutaneous ADRs, in health forum posts. Additionally, we discovered the novel ADR of hypohidrosis reported by 23 patients in erlotinib-related posts; this ADR was absent from 15 years of literature on this medication and we recently reported the finding in a clinical oncology journal. CONCLUSIONS: Several hundred million patients report health concerns in social health networks, yet this information is markedly underutilized for pharmacosurveillance. We demonstrated the ability of a natural language processing-based signal-generation pipeline to accurately detect patient reports of ADRs months in advance of literature reporting and the robustness of statistical analyses to validate system detections. Our findings suggest the important contributions that social health network data can play in contributing to more comprehensive and timely pharmacovigilance.