The power consumption of future user equipments (UEs) will be affected by the projected growth in their computing capacity, while data throughput may be affected by emerging aerial UEs with specific radio propagation conditions compared to terrestrial UEs. In such a context, this letter evaluates a key metric of interest, namely the probability that the uplink energy efficiency (EE) at a typical ground base station will be higher than a predefined threshold. We first characterize the priority bias of each UE layer as a function of long-term shadowing and system-level parameters to assess its penetration rate, i.e., the amount of active UEs from each tier among the total population of active UEs. Next, tractable approximations of the desired signal and the interference distribution are performed, enabling to derive the uplink EE. Our results demonstrate that an aggregation of the system-level parameters through the aerial priority bias needs to meet a given constraint to mitigate interference from aerial UEs and enhance the uplink EE of ground UEs. Monte-Carlo simulations validate the accuracy of our analytical results.