Although increasing community access to public schools through shared use agreements (SUAs) has been a recommended strategy for promoting physical activity (PA) among national, state and local organizations, empirical evidence examining the efficacy of SUAs is limited. This study examined the degree of usage and production of PA among schools with shared use, and how variation in PA output is related to characteristics of the school, type of activity, facility type, and when activity occurs. Data were collected in 20 schools across North Carolina using System for Observing Play and Recreation in Communities (SOPARC) and Structured Physical Activity Surveys (SPAS) to assess PA in school athletic facilities during out of school time. Findings indicated that although schools had a policy of shared or open use, most facilities were empty during non-school hours. Hierarchal linear regression models also showed that formal programming was positively associated with both use and PA levels. Given the abundance of empty facilities, community groups in need of space to facilitate structured PA programs should pursue avenues of sharing facilities with public schools. Furthermore, to increase the efficacy of shared use, structured physical activity programs may be needed. Future studies are encouraged to further explore the effects of the specific types of shared use programs on PA production as well other aspects of the built environment surrounding schools.