|
|
|
Index - Major Sections
Site Map
Product and Services _______________ Index - Same Level Subject
Index - Child Subjects |
IntroductionData is static. It doesn't change and it is never out-dated...Data is just a "oh yeah" type of thing. Unless data is compared with other data over time it has no meaning...Even data that answers a "yes/No" Question requires the Question to be given so the user knows what context the yes or no answer was given. Data by itself gives a manager nothing he can used to make decisions...It is only by analyzing data that data becomes information. In the past, data was collected using a standard procedure. First, a "Need of Information" survey was taken from each and every unit within an organization to determine their requires for information. Without any surprise, each unit had different needs (and many had no idea what they needed)
After these needs were itemized, "indicators" and summary reports were developed. This data was then prioritized and "limited" to only what was considered absolutely necessary because you didn't want to "distract" the user with too much information. For lower level users it was consider sufficient to provide them with a few key indicators in order to manager.
Finally, it was necessary to program separate applications for each of the final requirements for each of the intended users.
Although much time was spend in giving the units "what they wanted," because of the requirement to prioritize and to limit the collected data, there was always something missing. When any design had to be change to add more variable it was almost impossible to do and then it was expensive and time consuming.
The rule was "one a year or so" reevaluations were made and the procedures changed. No effort was made to make corrections as they were necessary. Time and money was wasted. Most informational systems at the clinical level, store data in multiple files, index cards, or other types of registers. Generally, each and every “statistics” that is kept, is kept separate. Data across records are duplicated and no data integrity ( the same information in different records is not the same). There are no links that connects one record with another (same as in the present US system). These organizations create reams of data, which is regulated to a room filled with more reams of paper never to be seen again. In some cases the paper and the reports completely fills work areas with only a small corner of the room set up with tables where the staff actually are able to works. Very little data is kept on any one patient—only
summary statistics are kept by type of event, usually in separate registers.
No quality or continuity of care for any one client. Individual researchers come and go, collect their data, and then disappear with "their data" (yet it was probably paid for with tax payers money) that is never seen again. Again, time and money was wasted, but more importantly, valuable information was lost. It is the purpose of this Web site to propose the use of Clinical Base Computer Systems for all three purposes, operational, management and for research. It doesn’t even have to be tested, businesses have been using it for years with success. Integration of DataIncorporate data from multiple sources, translate different transactional systems information into an integrated data source, a common metadata dictionary, and common definitions. (Microsoft Integration Services©) The creation of an information system using Common Clinical applications, Web Technology, a Centralized Warehouse, and Data Mining allows a completely integrated system Integration of InformationThere is the ability to integrate all sources of
information into one system.
Health Care Health Care has the potential to generate huge volumes of data from family planning centers, mother and child health centers, clinics, hospitals, and other health care organizations. Yet much of this data is still not collected, or if it collected, is still processed manually in spite the successful application of information technology in other information-intensive industries (Grimson, Grimson and Hasselbring, June 2000). Traditionally, research data was collected, analyzed, put on a floppy disk and stored in a desk….never to be used again. The data that was collected was collected in flat files that were designed only to answer the particular question that a researcher had in mine. Once the system was designed to answer this particular questions, it was very difficult or impossible to view or query the data any differently. by the time the data was collected it was old and it was useless to use to make decisions. Imagine a business making a decision based on data that was "collected once a year." Real world situations require that raw data be able to turn into up-to-the-minute information necessary to answer new questions about new situations…or better yet, to be able to monitor real life situations. One must be able to take collected data and use that data to manage Data types
Reasons for data collectionThe degree of detail that should be incorporated into a database depends on the information required.
The conventional methods of data collection such as censuses, surveys, vital or institutional morbidity and mortality statistics are no longer sufficient in themselves to give Healthcare Professionals the information they need to manage health care.
The utility of data is unquestioned. But how does the utility present itself? Utility presents itself in the form a of a model. If I can describe the operation of natural phenomena with a few well-chosen data elements, then I can present a simple data summarization--a model built from data--that is easy to grasp and conceptualize (de Ville 2001).
Management requires detail data. Management needs to know why a particular indicator is the value that it has. By using models, an intelligence guess can be made on how to affect the results of the indicator...not just knowing its value. By manipulating the models we become active participants in the real world... instead of establishing a project and passively waiting to see what happens.
There are three separate reasons to collect information in a Health Care Organization
Operational data is data that is collected during the regular day-to-day events of the organization. Today, this data is commonly collected into registers and summarized to be forwarded to some higher level government agency. This data may be used for financial reporting, external reporting, or for monitoring and evaluation.
Transactions are representations of events. In order to construct a transaction, we must have both data and the relationships among the data. These relationships th
Data Needs
Complexity
Due to the complexity of human health systems, health care data also tends to be very complex. Each culture has their own ideas of what health care is and should be. Each clinician has his own way of writing even the simplest of data values and notes.
It is acknowledge that it is impossible to capture all the meaning of words written (the spoken language is as bad) and lost will occur in any attempt to simplify its structure. But it is also acknowledge that without this simplification and organization, it is impossible to make any meaning out of the data.
In order for data to import knowledge, data must be collected into a meaningful form. It must be organized, classified, and categorized.
There must be a trade-off between these seeming conflicting concepts. A good Information System must be able to balance this conflict.
Certainty and precision
Data of all types when recorded carry with them degrees of uncertainty. This relates to all information, but especially to clinical observations and interpretations. Although in most cases, certainty is not a problem in health care (users understand the "nature of the data") in other cases certainty may be life-treating.
The assessment of severity or risk in a situation is often as important as the recording of findings and may be the sole basis for management decisions. The triaging of a client to an incorrect classification can cost the client their lives.
A clear understanding of How and to What Use data is to be used is important in any data collection system.
Data Integrity
Organizations and environments change. Therefore the data and relationships must be capable of being altered or the data will become outdated and inaccurate.
Data Format
The data must be in a format that is conducive to performing sophisticated modeling and statistical analysis as well as run-of-the-mill query and analysis operations. Data must also be made available in a form that can be made available to stakeholders throughout the health care system. Data Collection Techniques (non-computerized)
Data Collection Issues
In public health a very high level of detail, granularity, is desired. There is a need to treat individual clients not "populations."
Databases
The whole purpose of a database is the representation of the physical world. The accuracy of the representation depends on the design of the database system.
|
|
|