The study examines the relationship between the impact severity of road accidents and the spatial grouping of Malaysian states during the COVID-19 pandemic in 2020 and 2021, addressing the lack of analysis on these trends in Malaysia. The road accidents impact severity is divided into two subgroups, which are injury severity level (i.e., death, bad injury, and light injury) and accident severity category (i.e., fatal, serious, and minor accidents). Nine combinations of these severity levels and categories for each observed year are analyzed using linear regression analysis. The significance and correlation for each combination is observed and studied. Further analysis is performed using quadrant analysis, grouping states based on the mean-centered positions. Subsequently, chi-square analysis is applied to these quadrant-based groups to explore associations between the spatial distribution of states and the combinations of severity levels and categories. The study aims to explore the spatial trends of road accidents in Malaysia, highlighting how accident and injury severities are distributed across different states during the pandemic. The findings showed that only the accident severity categories directly linked to the injury severity level have a strong linear relationship, regardless of the year. Pulau Pinang demonstrated a worsening trend, shifting from Quadrant III in 2020 to Quadrant I in 2021 across all combinations, while Selangor remained in Quadrant I for all cases throughout the pandemic. The impact severity for death and serious accident, as well as for bad injury and fatal accident, showed no significant relationships between the years 2020 and 2021.