Undermining support for COVID-19 public health policies: an analysis of the impact of subversive conspiracy narratives on Czech General Practitioners.

BACKGROUND: Limited knowledge exists regarding the impact of COVID-19 conspiracy theories on the professional practice of general practitioners (GPs). This study aimed to identify the basic characteristics of GPs who endorse COVID-19 conspiracy beliefs and compare their level of support for COVID-19 health policies with GPs who do not believe in conspiracies. METHODS: Between January and February 2021, a representative online survey was conducted among 1163 GPs in the Czech Republic. The sample was designed to be representative of members of The Association of GPs of the Czech Republic. RESULTS: The survey revealed that nearly 14% of the GPs surveyed believed in one or more COVID-19 conspiracies. The average age of GPs who endorsed conspiracies was 58, which was higher than the rest of the sample (average age of 50). GPs who believed in conspiracies were less likely to support COVID-19 public health policies and therapy recommendations, including vaccination. Logistic and linear regression analyses indicated that doctors who believed in conspiracies were 2.62 times less likely to have received a COVID-19 vaccine. Mediation analysis showed that approximately one-quarter (23.21%) of the total effect of trust in government information on support for public health policies was indirectly mediated by the endorsement of COVID-19 conspiracy beliefs. CONCLUSIONS: The study findings suggest a concerning association between belief in COVID-19 conspiracies and a reduced level of support for public health policies among GPs. These results underscore the importance of incorporating the 'conspiracy agenda' into medical authorities' more effective public health communication strategies.

CsFEVER and CTKFacts: acquiring Czech data for fact verification.

In this paper, we examine several methods of acquiring Czech data for automated fact-checking, which is a task commonly modeled as a classification of textual claim veracity w.r.t. a corpus of trusted ground truths. We attempt to collect sets of data in form of a factual claim, evidence within the ground truth corpus, and its veracity label (supported, refuted or not enough info). As a first attempt, we generate a Czech version of the large-scale FEVER dataset built on top of Wikipedia corpus. We take a hybrid approach of machine translation and document alignment; the approach and the tools we provide can be easily applied to other languages. We discuss its weaknesses, propose a future strategy for their mitigation and publish the 127k resulting translations, as well as a version of such dataset reliably applicable for the Natural Language Inference task-the CsFEVER-NLI. Furthermore, we collect a novel dataset of 3,097 claims, which is annotated using the corpus of 2.2 M articles of Czech News Agency. We present an extended dataset annotation methodology based on the FEVER approach, and, as the underlying corpus is proprietary, we also publish a standalone version of the dataset for the task of Natural Language Inference we call CTKFactsNLI. We analyze both acquired datasets for spurious cues-annotation patterns leading to model overfitting. CTKFacts is further examined for inter-annotator agreement, thoroughly cleaned, and a typology of common annotator errors is extracted. Finally, we provide baseline models for all stages of the fact-checking pipeline and publish the NLI datasets, as well as our annotation platform and other experimental data.