There is a strong motivation, in particular in the domain of healthcare, for new perspectives on context driven research and computing in order to provide next generation services to people that are tailored to individual needs rather than generalised assumptions that could potentially endanger human life. For that, context-awareness is a key requirement in order to reach a better understanding of human-centric computing systems and environments and subsequently, the deployment of dedicated services that are specifically adapted to the context to which they are applied. Such context-driven services would be able to provide the means of delivering situation-aware and person centric services that ultimately may even anticipate future behaviour and problems of the user itself and the context in which the user finds themselves. However, the perpetual provision of contextual data in pervasive environments is far from being easy and includes major challenges that vary between environments. The reason for this is not only the sensor diversity within the environments themselves but also the contextual scope to be analysed and the amount of data to be collected and correlated to actually reach a minimum degree of contextual understanding. For that reason, in this paper, contextual environments have been categorised as well as their interaction into different groups that reflect individual contextual levels of interest of which contextual understanding is required and consequently to which services can be applied.