Abstract Background Tailoring pre-exposure prophylaxis (PrEP) service delivery is key to scaling-up PrEP uptake. Optimal implementation of tailored services requires, among other things, insights into patterns of PrEP use, sexual behaviours and condom use over time. Methods Between September 2020 and January 2022, we conducted a web-based, longitudinal study among PrEP users in Belgium. In three questionnaire rounds every six-months, we assessed PrEP and condom use, and sex with steady, casual and anonymous partners in the preceding three months. Based on the patterns of PrEP use in the preceding three months, we identified distinct PrEP use categories. We investigated differences in baseline socio-demographics and sexual behaviours by PrEP use category using Fisher’s exact and one-way ANOVA tests. Patterns in PrEP and condom use over time were examined using descriptive analyses and visualised in alluvial diagrams. Results In total, 326 participants completed the baseline questionnaire, and 173 completed all three questionnaires. We identified five distinct PrEP use categories: daily (≥ 90 pills), almost daily (75–89 pills), long period (> 7 consecutive days and < 75 pills) with or without additional short period use, short period (1–7 consecutive days and < 75 pills) and no PrEP use (0 pills). During the study, percentages of individuals in each PrEP use category varied, but did not change significantly over time. At baseline, daily and almost daily users were more likely to report five or more casual sex partners, ten or more anonymous sex partners and anal sex on a weekly basis with casual or anonymous partners compared to those using PrEP for long or short periods. Up to 12.6% (n = 16/127) of participants reporting anal sex with casual or anonymous partners, indicated always using condoms and PrEP with these partners. One in three (n = 23/69) participants who reported anal sex with steady partners had condomless anal sex and did not use PrEP with these partners; with casual or anonymous partners ...