With the development of wireless communication technology, wireless body area networks (WBANs) have become a fundamental support tool in medical applications. In a real hospital scenario, however, the interference between wireless medical devices and WBANs may cause a high packet drop rate and high latency, which is harmful to patients using healthcare services. Nonetheless, cognitive radio is a promising technology for sharing the precious spectrum, which has high efficiency of the wireless resource. Thus, WBANs with cognitive radio capability are also exploited. We propose a spectrum-aware priority-based link scheduling (SPLS) algorithm for cognitive radio body area networks (CRBANs) in a real hospital scenario. In SPLS, three channels are used: DataCh, EDataCh, and CtrlCh for normal data, emergency data, and control messages, respectively. To avoid collision during data transmission, neighboring CRBANs send messages regarding the channel state with CtrlCh before the scheduling. The CRBANs can share DataCh in the time domain for improving the throughput. The SPLS algorithm allows a CRBAN to access idle channels on the licensed and unlicensed spectrum according to the CRBAN traffic. Our simulation results show that the proposed SPLS outperformed the conventional scheme in terms of packet delivery ratio, system throughput, latency, and energy efficiency.