Introduction: Spatio-temporal parameters (STP), calculated from 3D gait analysis, are frequently used for treatment planning and evaluation in Cerebral Palsy (CP). To calculate these parameters, accurate determination of gait events (i. e. initial contact (IC) and foot off (FO)) is essential. Previous research on the performance of kinematic gait event detection algorithms on different walking patterns led to recommendations, which have not been verified on clinical populations.  Research questions: 1) Which current kinematic approach is best capable of determining IC and FO for diverse gait patterns? 2) Does the use of automated kinematic algorithms affect clinical interpretation of STP compared to current clinical event detection (force-plate, visual identification)? Methods: 3D kinematic and kinetic data was retrospectively collected from 90 children with CP. Participants were classified in 3 categories – groups A (fore-foot IC), B (flat foot IC) and C (heel IC). Five kinematic algorithms (one modified) were implemented for two different foot marker configurations for both IC and FO and compared with clinical (visual and force-plate) identification using Bland-Altman analysis. The best-performing algorithm-marker configuration was used to compute STP, which were compared with those obtained clinically. Results: In agreement with previous studies, sagittal velocity of the heel (Group C) or toe markers (Group A and B) was the most reliable indicator of IC, and the speed-dependent sagittal velocity coupled with the hallux marker worked best for FO across the entire dataset. A comparison of kinematic and clinical showed >1.78% differences in spatial parameters, and >6.3% differences in temporal parameters. Significance: Outcomes showed that the choice of the best-performing algorithm was dependent on a combination of algorithm and marker choice. However, observing the high differences between clinical and kinematically calculated spatio-temporal parameters, clinicians need to be aware that the differences could likely affect clinical interpretation of gait analysis results. Hence, further research is needed to establish the efficacy of implementing automatic gait event detection algorithms in a clinical setting.