Correcting misconceptions and false beliefs is important for inserting reliable information about COVID-19 into public discourse, but what impact does this have on the continued proliferation of misinforming claims? How can we track their impact over time? What is the best way to inform individuals about the misinformation they share? Using more than 3 years of data collected from Twitter and fact-checking organisations, we discuss the relation between fact-checking and misinformation across topics and demographics. We then proceed to show how the Fact-checking Observatory, a website that generates human-readable weekly reports automatically about the spread of covid-related misinformation and fact-checks can be used for monitoring such information over time. Finally, we analyse early results about the effectiveness of our Twitter bot in reducing individual sharing of misinforming content.