Purpose: Telemedicine systems must provide clinical data of sufficient quality (according to medical standards) to support safe treatment guidance of outpatients. Quality of clinical data (QoD) typically varies due to unstable performance of ICT-components of these telemedicine systems. Therefore, telemedicine systems that support treatment guidance of outpatients should be QoD-aware and must be able to appropriately adapt treatment guidance to QoD variations. Only in this way, the effectiveness of treatment guidance and the safety of outpatients can be guaranteed. Methods: This paper follows a design science approach for the development of a functional architecture for QoD-aware telemedicine systems, with emphasis on one key component: the QoD Broker. Existing requirements elicitation methods were refined to deal with capturing QoD-specific requirements. Furthermore, an ontology-driven knowledge management method is proposed to enable the correct interpretation and manipulation of QoD within telemedicine systems. The functional architecture was validated using various methods, including prototype experiments and expert interviews. Result: One of the key components of the proposed functional architecture is the QoD Broker, which is a novel component that adds QoDawareness to telemedicine systems. The QoD Broker obtains quality of service data from ICT components within a given telemedicine system and uses different computational models to compute QoD. This paper presents the QoD Broker architecture and the QoD management techniques implemented in the QoD Broker: (1) QoD dimensions, (2) QoD evaluation (i.e. computational models), (3) QoD stratification models and (4) technological recommendations. Discussion: The paper presents partial results of the validation that was performed in the context of the European MobiGuide project. The validation confirms that the proposed QoD Broker satisfies stakeholders’ requirements and is considered useful to support stakeholders’ goals.