As mentioned above, EAV modeling makes the meaning of data categories, such as clinical findings, numerous and sparse. If these conditions are not true, the standard relational model (that is, each attribute column) is desirable; Using EAV does not mean giving up common sense or the principle of good relationship design. In clinical record system, subschemas usually imitates tradition to deal with patient's demographics and bills. (Although most suppliers' database models are proprietary, VISTA, the whole system of the US Department of Veterans Affairs (VA) medical system, is called the Veterans Health Administration (VHA )[ 1]. It is open source and its model is easy to check, although it uses mumps database engine instead of relational database. )
As will be mentioned later, EAV database is non-maintainable metadata, and there are basically no related tables, including support. Metadata tables, usually compared with EAV tables, have at least three or more factors, usually standard relational tables. [2] [3] An example of the metadata table is the attribute definition table mentioned above.