Objective: To investigate whether undertaking a physical activity- level intervention influences subsequent participation attendance (the frequency or range of activities undertaken) and involvement (the experience of participation while attending) in physical activities for children with cerebral palsy. Design: Systematic review and narrative synthesis following the Preferred Reporting Items for Systematic Reviews and Meta- Analyses guidelines. Method: Published literature from 2001 to April 2019 was searched within six databases, with hand-searching of reference lists of included articles. Included studies were: intervention studies of participants aged 7-18 years, when ≥50% had cerebral palsy; where participation attendance and/or involvement outcome/s in any physical activity were measured; and follow up was ≥1 month beyond the intervention period. Study selection, data extraction and assessment of risk of bias was completed independently by paired reviewers. A narrative synthesis describes the findings. Results: Twelve studies were included; 10 randomised controlled trials (RCT) and two case studies. Study quality varied from low to high (RCT PEDro scores: 5-10). Participants in included studies were 4.0-16.7 years; sample sizes were n=6- 34 in intervention groups (n=444 total participants). Physical activity interventions varied in: being individual or group-based; the types of exercises, intensity and load; with two focusing on the upper limb, two on treadmill training, four on functional resistance training/fitness, one on specific physical-activity goals and one multi-strategy approach. Interventions varied in frequency (1-3 times/week), duration (30- 60 minutes), length of intervention programme (10 days- 8 months) and follow up time (1 month-6 months). Participation was the primary outcome in four studies; 11 reported on attendance and two on involvement. At follow-up, one study demonstrated between-group differences in frequency of attendance favouring a multi-strategy intervention approach over standard care, five demonstrated no differences, and one favoured the comparison group for improved diversity of participation. In studies comparing two physical activity interventions, one demonstrated increased attendance following high frequency compared to low frequency interval treadmill training, and one following supported treadmill training compared to home-based strengthening. Two studies reported no differences between interventions. One case study reported positive changes at follow up. Conclusion: Physical activity interventions may influence participation attendance; but not all gains are sustained. The participation construct, involvement, remains under researched and requires further attention. Further research is required to determine the most effective physical activity interventions needed to enhance ongoing participation.