Abstract Objectives Rituximab (RTX) is a safe and effective treatment for RA. However, there are some concerns about infection risk and preliminary data suggest dose and time dependency. This study aims to determine the infection incidence in a large real-life population of RA patients using RTX, with special focus on (ultra-)low dosing and time since last infusion. Methods RA patients treated with 1000, 500 or 200 mg RTX per cycle between 2012 and 2021 at the Sint Maartenskliniek were included in a retrospective cohort study. Patient-, disease-, treatment- and infection characteristics were retrieved from electronic health records. Infection incidence rates, dose and time relations with RTX infusion were analysed using mixed-effects Poisson regression. Results Among 490 patients, we identified 819 infections in 1254 patient years. Most infections were mild and respiratory tract infections were most common. Infection incidence rates were 41, 54 and 71 per 100 patient years for doses of 200, 500 and 1000 mg. Incidence rate ratio (IRR) was significantly lower for 200 mg compared with 1000 mg (adjusted IRR 0.35, 95% CI 0.17, 0.72, P = 0.004). In patients receiving 1000 or 500 mg RTX, infections seemed to occur more frequently within the first two months after infusion compared with later on in the treatment cycle, suggesting an association with peak concentration. Conclusion Ultra-low dosing (200 mg) of RTX is associated with a lower risk of infections in RA. Future interventions focusing on ultra-low dosing and slow release of RTX (e.g. by subcutaneous administration) may lower infection risk.