The COVID-19 pandemic has driven the transition from face-to-face visits to virtual care delivery. In this study, we explore patients’ perceptions of the benefits and challenges of using virtual primary care technologies during the pandemic, using machine learning approaches. A cross-sectional survey was conducted in August 2020 in Italy, Sweden, Germany, and the UK. Latent Dirichlet Allocation was used to identify themes of two open-ended questions. Comparisons between participant characteristics were made using Wilcoxon rank-sum test. 6,331 participants were included (51.7% female; 42.4% +55 years; 60.5% white ethnicity; 86.6% low literacy). The benefits extracted included: primary care delivery, infection control, reducing contacts, virtual care, timeliness, patient-doctor interaction, convenience, and safety. Participants from Sweden were most likely to mention “primary care delivery” (UK p = .007, IT p = .03, DE p < .001), from the UK “virtual care” (SE p < .001, IT p < .001, DE p < .001) and from Italy “patient-doctor interaction” (UK p < .001, SE p < .001, DE p < .001). The challenges included: diagnostic difficulties, physical examination, digital health risks, technical challenges, virtual care, data security and protection, and lack of personal contact. “Diagnostic difficulties” was most significantly mentioned in Sweden (UK p = .009, IT p < .001, DE p < .001), “virtual care” in the UK (IT p = .02, SE p = .001, DE p < .001), and “data security and protection” in Germany (UK p < .001, IT p = .019, SE p < .001). Our study reinforces the feasibility of using machine learning to explore large qualitative datasets. Our findings contribute to a better identification of the lessons learned during the pandemic and inform improvements in policy and practice.