At the beginning of the coronavirus disease (COVID-19) pandemic, hydroxychloroquine (HCQ) was widely used as a possible antiviral agent. Current knowledge indicates that HCQ has little or no effect on individual clinical outcomes of COVID-19, but populational effects on disease transmissibility are still unknown. This study investigates the hypothesis that massive HCQ consumption by a population may contribute to reducing the transmissibility of SARS-CoV-2 and COVID-19 spread by reducing the viral load of infected individuals. Public database of seven states from Brazil in 2020 were assessed, before the start of COVID-19 vaccination. The daily values of the COVID-19 effective reproduction number (Rt) were obtained. Associations between Rt values and the proposed predictor variables (prevalence of COVID-19 as a marker of collective immunity; social isolation indices; consumption of HCQ) were tested using multiple linear regression analysis. In all seven states, consumption of HCQ was a significant negative predictor of Rt (β ranged from -0.295 to -0.502, p = 0.001). Furthermore, the mean derivative of Rt during the declining period of the COVID-19 incidence (the mean rate of variation) was also significantly negatively related to the mean HCQ consumption in that period (R2 = 0.895; β = -0.783; p = 0.011), meaning that the higher the HCQ consumption, the faster the decline of COVID-19 Rt. It suggests a dose-response phenomenon and a causal relationship in this association. The results of this study are compatible with the hypothesis that HCQ has small but significant in vivo antiviral effects that are able to reduce SARS-CoV-2 transmissibility at the populational level.