Time is one of the most difficult aspects to handle in real world applications, especially in database systems. Relational database management systems proposed by Codd offer very little built-in support for managing time varying data and theory of temporal semantics. Many temporal extensions of the relational model have been proposed and some of them are also implemented. The proposed system is based on an ontology of health Care terminology including patients vital data, activities and as well as the treatment data. We will propose a conceptual temporal model for handling time varying attributes in the relational database model with minimal temporal attributes. The proposed model is easy to define, manage and incorporates the important and relevant features in the target temporal relational model. Furthermore we have illustrated implementation of the model on patient database and also present the requirement analysis: work flow for managing patients in hospital environment. In response to rising health care costs, reliability, privacy, security and changing expectations concerning the quality of health care, data management is very important in improving health care services.
The relational model  is based on a brand of mathematics called relational algebra. Codd used that concept to manage huge amounts of data very effectively. Codd and others have extended the notion to apply to database design. Thus they were able to take advantage of the power of mathematical abstraction and the expressiveness of mathematical notation to develop a simple but powerful structure for databases .
A relation has to be in first normal form (FNF), meaning that the domains of the attributes in its schema may only be of scalar data types. In other words, a relation can be considered as a subset of the Cartesian product of all the attribute domains contained in its schema.
The relational data model only support functionality to access a single state of the real world, called a snapshot. The transition from one database state to another (updates) thereby giving up the old state. There exist, however, many application domains which need to have access not only to the most recent state, but also to past and even future states, and the notion of data consistency must be extended to cover all of these database states. Due to the FNF assumption in the relational model, there is a restriction in expressing the data structures. To overcome this drawback, the relational model has been extended  to support, non first normal form (NFNF) or nested relations.
Efforts to incorporate the temporal domain into database management system have been ongoing for more than a decade and dozens of temporal models have been proposed , , ,  and a few of them have been implemented , , , .
Designing effective, secure and useful healthcare information systems which handles temporal data is a great challenge for software engineers. It includes very complex information that evolves with time. Relational model is a very powerful model and well accepted model among the vendors. There are number of extensions to this model which incorporates time varying data. In this paper we investigate the patient data management (PDM) with respect to the time varying nature of data and propose a conceptual temporal relational schema for PDM.
This paper is organized as follows: Section2, discussion on various temporal relational models has been made. Section 3 deals with the ontology and a proposed requirement analysis: work flow for patient data management. Section 4 describes the proposed conceptual data model , (TempR-PDM) with its logical schema.
Many data models are introduced so far to capture the semantics of temporal data keeping the traditional entity relationship model (ERM) approach. Traditional ERM is not capable for capturing the whole temporal aspects. Many extensions  have been proposed to extend the ERM to capture time varying information in one way or the other. Unified modeling language (UML) is also used as a tool to develop the logical and conceptual schema of the mini world.
The other important point is how this new conceptual model  will be incorporated into a relational database. One way of doing this is to develop a temporal layer and this layer is responsible for translating the temporal queries to traditional SQL statements. The other approach is to design a complete temporal query language  which not only supports all SQL statements but incorporate new operators based on temporal relational algebra . There are many solutions for this problem and few successful implementations are also summarized in table 1.