Background Successful stroke trials require adequate recruitment. In this observational study, we assessed reasons for refusal to provide informed consent in eligible patients approached for clinical trial participation at the Vancouver Stroke Program. Methods We assessed screening logs from four trials that were actively recruiting at our center: three randomized trials, two of which investigated different antithrombotic strategies for secondary prevention (NAVIGATE-ESUS, NCT02313909 12/2014; DATAS-II, NCT02295826 11/2014) and one that investigated surgery plus medical management versus medical management alone for primary prevention (CREST-2, NCT02089217 03/2014). The fourth study was observational and non-randomized; all participants received an external monitoring device (PROPHECY, NCT03712865 10/2018). Screening logs from June 2015 to April 2017 were reviewed retrospectively. Subsequently, we used a prospective structured case report form for screening (May 2017–March 2018). We assessed and compared refusal rates between trials, demographics of those refusing consent, and their reasons for doing so. We used descriptive statistics, chi-square and Fisher’s exact tests as appropriate for non-parametric data, and t-tests for parametric data. We examined likelihood of refusal by sex using multivariable logistic regression models including age and trial intervention as co-variables. Results A total of 235 patients (43% women) were approached for consent. More patients refused the surgical (59%) and antithrombotic trials (53%) compared with the non-randomized external monitoring device study (13%) ( p < 0.001). Surgical trial refusals were primarily due to a desire for certainty in receiving a particular intervention (39%), with the majority of those patients wanting surgery. Refusals for the antithrombotic trials were mainly due to concerns with the potential side effects of the study drug (41%); refusals in the device trial were mainly due to disinterest (46%). Women refused participation more often than men (48% vs 33%). Women remained less likely to consent than men, even after adjustment for age and trial intervention (OR 0.46, 95% CI 0.26–0.82, p = 0.009). Conclusions Concern surrounding drug safety, randomization, and disinterest were the chief deterrents to enrolment; there were also differences in rates of consent by gender. A better understanding of why patients refuse participation in stroke trials may help to develop future patient-directed communication strategies to improve enrolment. Further research is required to better understand the reasons underlying gender disparities in consent rates. Electronic supplementary material The online version of this article (10.1186/s13063-019-3434-0) contains supplementary material, which is available to authorized users.