Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2000-10-19
2004-12-07
Corrielus, Jean M. (Department: 2172)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000
Reexamination Certificate
active
06829604
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to the data processing field, and more specifically, to a method and system for evaluating the efficiency of and reordering accordingly a plurality of exact and probabilistic enterprise search rules.
2. Description of the Related Art
As healthcare organizations strive to provide maximum value to their customers, access to comprehensive patient information is more important than ever before. The rampant consolidation that has resulted from mergers and acquisitions has made it imperative that providers be able to track a patient across multiple facilities and throughout multiple episodes of care. In many healthcare systems, multiple systems within the same organization are maintained individually resulting in duplicate records of the same person causing confusion if an individual is not entered with exactly the same identification information at each entry point. This task of linking a patient across disparate information systems that are not integrated is a formidable challenge for most organizations; the duplication of critical patient information records caused by manual data entry makes the challenge greater.
For these reasons, an accurate method of identifying individual persons within enterprise is the critical foundation for the healthcare system, so it is essential that master patient identifier systems provide certainty that users are acting upon patient information that is complete, accurate, and updated—that the correct test result is linked to the correct patient. The effects of mistakenly identifying a patient in a healthcare enterprise can be far-reaching, whether the organization comprises one facility or a growing integrated health network. Unnecessary resource utilization, customer dissatisfaction, billing discrepancies and even the possibility of legal exposure from erroneous clinical decisions based on inaccurate patient information—all can be the unfortunate repercussions of maintaining disparate demographic, clinical and financial information about a person that cannot be linked across the enterprise.
Master person index systems are not new, but have never fully addressed the complexities of clearly identifying an individual in today's multi-layered healthcare environment. Some such systems are functionally capable of producing printed reports, others can function with same-vendor systems, and still others rely on hard-programmed matching criteria with limited flexibility. While most master person index systems have been exclusively patient-based or member-based, or have been designed to link systems from a single vendor, effective healthcare delivery in today's environment calls for a more comprehensive solution.
Many master person indexes utilize a series of pre-defined rules to search for the desired target patient records. The rules typically are comprised of a series of elements or fields that enable searching for the desired individual. For example, a rule may include a person's last name, social security number, telephone number, and zip code. Thus, in this example, a user may attempt to find a particular patient by entering the patient's information, and this sample rule would utilize the patient's last name, social security number, telephone number, and zip code. In executing the rule, the system would search the patient database for the same last name, social security number, telephone number, and zip code. The rule would record a hit if it found exact matches for these four data items, but the rule may also record a hit if it found a close match as well. However, if this rule does not find a record that closely matches the data elements for these four data items, no record would be returned and the rule would fail.
Because a given rule may not, and likely will not, always find the desired patient, master person index systems utilize many rules comprised of various data element combinations. In fact, some rules may be implemented in a system that rarely if ever find matches due to the construction of the rule. Therefore, upon firing many different rules, the likelihood of retrieving the correct patient record greatly increases. However, master person index systems create new problems in maintaining many rules and attempting to consider the large number of results that are actually false hits.
The problem with defining rules to locate specific individuals in the enterprise results in situations where some rules are more efficient in finding the correct result that others, but if the good rules are not fired first or before the less-efficient rules, the correct record may be buried in a long list of potential records for the desired individual. Moreover, some rules to locate records may take longer to execute to return the results. Yet other rules may not actually fire if their order is such that the desired number of results is reached by the higher ranking rules. In this situation, the higher ranking rules may not return the correct result while the unfired rule may have. In addition, some rules may have a higher rate of returning duplicate records than other rules, so unless those rules are identified and either adjusted or deleted, their existence reduces system efficiency. Even still, other rules may create a high number of false hits thereby increasing the operator's time in determining the proper record. Finally, rules may be misfired because of data entry errors. Thus, there is a problem in maintaining a set of rules to locate the desired records in an efficient manner.
SUMMARY OF THE INVENTION
A rules analyzer system and method is provided for an enterprise system to evaluate and rank exact and probabilistic search rules for searching a computer database of records according to the efficiency of each search rule. The rules analyzer collects statistics on the performance of each search rule and assigns a priority value for each search rule according to the collected statistics. The priority values are based on the efficiency or precision of each search rule. Thereafter, the rules analyzer ranks the search rules according to the assigned priority.
In another embodiment, system for uniquely identifying an object record in a database of object records according to a plurality of exact and probabilistic search rules is provided. The system is configured to receive application identification information and attributes that correspond to a target object contained in the database of object records. The system additionally executes one or more exact-match search rules to search the database of object records for the target object. If the exact-match search rules do not return the target record, the system executes one or more user defined probabilistic search rules to search the database of object records for the target object. A list of probable matches to the target object is retrieved by the probabilistic search rules, and the probable matches are ranked by degree of match probability. The user of the system thereafter determines which retrieved record is the target record. Upon selection, the system updates the database of object records in real time for the selected target object with new attributes and information associated with the target object. The system determines the efficiency of the exact-match and probabilistic search rules according to collected statistics for each search rule. Finally the rank order of the exact-match and probabilistic search rules are rearranged in descending order by the efficiency of each search rule.
Other systems, methods, features, and advantages of the present invention will be or become apparent to one with skill in the art upon examination of the following drawings and detailed description. It is intended that all such additional systems, methods, features, and advantages be included within this description, be within the scope of the present invention, and be protected by the accompanying claims.
REFERENCES:
patent: 4752890 (1988-06-01), Natarajan et al.
patent: 5204939 (1993-04-01), Yama
Corrielus Jean M.
Eclipsys Corporation
Liang Gwen
Thomas Kayden Horstemeyer & Risley LLP
LandOfFree
Rules analyzer system and method for evaluating and ranking... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Rules analyzer system and method for evaluating and ranking..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Rules analyzer system and method for evaluating and ranking... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3336044