This paper studies the under-reporting of workplace injuries in Italy, leveraging administrative data on work accidents. Using a difference-in-differences approach, I compare injury reporting behavior across provincial economic sectors exposed to the news of a fatal workplace accident with those not exposed, providing a causal estimate of the effect of the news. The analysis reveals that, in the weeks following such news, the weekly number of non-severe injuries reported per 100,000 workers increases. This suggests that, under typical conditions, a substantial number of non-severe accidents remain unreported. Two mechanisms drive this pattern: media coverage, which likely puts pressure on both firms and workers to better comply with reporting standards, and a strong presence of workers’ unions. In fact, a decomposition of the ATT shows that the effect is stronger in provinces with higher union membership.