There are seven health authorities and each of them is responsible for a number of hospitals. Currently each health authority maintains their own database system and they do not use same database management system. Some of them uses oracle, some MS SQL and others custom database management system. This report presents a case study of a Data Warehousing and Data Mining used in a Regional Health Authority (RHA), its main function and how those two data management technology can solve problems faced by the Regional Health Authority.
Below is a definition of data warehousing and data mining and their advantages and disadvantages.
I.1 Definition
Data warehousing
Data warehouse is combining data from multiple and usually varied sources into one comprehensive and easily manipulated database1.
Data warehouse is a subject-oriented, integrated, time varying, non-volatile collection of data in support of the management's decision-making process2.
Fig 1: Data warehouse (Hammergren. 2009. P. 12).
Data Mining
Data mining (sometimes called data or knowledge discovery) is the process of analyzing data from different perspectives and summarizing it into useful information, information that can be used to increase revenue, cuts costs, or both. Data mining software is one of a number of analytical tools for analyzing data. It allows users to analyze data from many different dimensions or angles, categorize it, and summarize the relationships identified. Technically, data mining is the process of finding correlations or patterns among dozens of fields in large relational databases.
Figure 2: Knowledge mining (Jiawei Han. 2006. P. 7).
I.2 Application of Data warehousing in Regional Health Authority
The Regional Health Authority is responsible for seven local health authorities. Each Health Authority designs and implements its own database for patient, staff, treatment data. If the hospitals and health authorities perform differently there is a way to solve this problem by implementing one centralized database which will contain all the hospital's data (patient information, doctor's information, nurse's information, and types of treatment for a specific disease applied on patient for cure). By using data warehouse and analyzing the information RHA may find appropriate treatments for disease which has been succeed in many hospitals. With the implementation of Data warehousing all the hospitals share the same centralized database, doctors, nurses have access of a lot of patient's information, type of treatment applied to any disease. The access and use of that information improve decision making of the doctors and nurses for example to cure diseases. Furthermore the data accessed through the data warehouse is cleaned and used by the health authority.
If there would be a data warehouse for all health authorities, Regional Health Authority would have a control over it and be able to force some policies and standard to the health authority so that all the hospitals follow the same policies. Once health authorities and hospitals start following the same policies and start treat patient based on the same policies, patient will be satisfied.
Data warehouse can play a major role in case all the hospitals build and share the same data warehouse. As we know the data warehouse contain the copy of data/information from various source of databases. The database of each health authority contains a vast amount of patient data which will be copied to the data warehouse accessible by all the hospitals. After building a data warehouse for Regional Health Authority, the data warehouse will house patient data that can be shared with other health authority. Doctors, nurses can obtain detailed information about a patient instantly. This will provide immediate feedback on a patient's condition, plan of care, and current medication. By accessing information, doctors and researchers may use some OLAP tools to analyze patient information for the research purposes and discover new treatments.
Definition of OLAP tools: Short for Online Analytical Processing, a category of software tools that provides analysis of data stored in a database. OLAP tools enable users to analyze different dimensions of multidimensional data. For example, it provides time series and trend analysis views. OLAP often is used in data mining.
If a data warehouse is implemented correctly it can be a platform that contains health authority patient information in a centralized and normalized form for deployment to doctors, nurses and other hospital's staff to enable them to perform simple reporting, complicated analysis, benchmarking, clinical protocol development.
Besides the advantages of implementing a data warehouse there are some disadvantages of data warehouse. The disadvantages of implementing a data warehouse is that data which come from multiple sources must be cleaned, loaded, or extracted before the data can be stored within the data warehouse. This process can take a long period of time. In the scenario of Regional Health Authority it will take a long period of time to doctors, nurses to get access to patient information. The information needed will not be available in time. Users who will work with the data warehouse of the Regional Health Authority must be trained to use this technology because if they are not trained properly they may chose not to work within the data warehouse.
There may also be issues with compatibility. For instance new transaction systems may not work with systems that are already being used. Furthermore users who will work with the data warehouse must be trained to use it. If they are not trained properly they may chose not to work within the data warehouse. If the data warehouse is accessible via the internet, this could lead to a large number of security issues and privacy. It means that patient information will be exposed to malicious person. A data warehouse can be costly to maintain. Once Regional Health Authority has invested for the data warehouse, the management of RHA will still need to pay for the cost of maintenance over time.
Below I drew a diagram of the data warehouse for the Regional Health Authority using Microsoft Office Visio 2007.
Figure 3. A diagram of the data warehouse for the Regional Health Authority
In this diagram we see that information comes from different health authority's databases around the Regional Health Authority. That information is patient information, staff information, treatment information. Before information is loaded into the data warehouse, it must be conditioned or cleaned to ensure that it is good. The data warehouse team removes bad data and resolves any inconsistencies in the data that comes from the operational systems.
After the information is cleaned, it is loaded into the data warehouse where it waits to be accessed. In the figure 3, doctors, nurses from different health authority can access patients' health information from data warehouse by creating requests for data with the oracle discover tool to access patient information.
I.1 Application of Data mining in Regional Health Authority
Regional health authority can use data mining technology to uncover hidden and unexpected patterns in data for strategic decision-making in healthcare. By accessing information, doctors and researchers may use some OLAP tools to analyze patient information for the research purposes and discover new treatments. If all the hospitals and health authorities perform differently, Regional Health Authority may suggest to set up same policies on hospitals. For example in the policy they can suggest that the length of stay as in patient should be the same in all the hospitals for the same condition. After every treatment doctors can make a report of treatment which will be stored to the data warehouse. Then in the future if doctors face any difficulties then they can refer to those reports to cure the disease. There is a lot of knowledge to be gained from computerized health records. Health Authority's doctors can use data mining technology to analyze , examine amount of information related to patient and make a good decision making for instance to cure disease. When medical institutions apply data mining on their existing data, they can discover new, useful and potentially life-saving knowledge that otherwise would have remained inert in their databases.3 After implementing data warehouse for all hospitals, other hospital's managements may use data mining technology to access and analyze information for example on campaign on immunization against influenza inside the data warehouse which was applied by one hospital and succeeded in order to implement it to other hospitals. By implementing data mining Regional Health Authority will be able to make decision on how hospitals and health authorities should operate.
Nonetheless data mining has some disadvantages too. One of those disadvantages is privacy. With the widespread use of internet patient information may be accessed by somebody and use that information in an unethical manner. There is also a security issue. Regional Health Authority has a lot of information about its patients. But they may not have sufficient security systems to protect patient information.
