Systems and methods for decoding payer identification in health care data records
View Patent ↗Processing arrangements and methods are provided for the automated decoding or translation of information in healthcare data records, which are coded in a non-standardized or varying formats. A data record which contains information, a portion which is recognized and another portion of which is new, is decoded or translated using a statistical mapping rule. The mapping rule assigns a most likely translation value to the information based on the recognized portion of the information. The statistical mapping rules are established by analysis of a set of previously decoded data records.
1. A method for establishing mapping rules for assigning codes to prescription transaction data records, wherein each data record comprises at least a first, a second and a third data field, the method comprising:
receiving a set of previously mapped data records in which codes have been assigned to the data records based on at least the values of the third data fields;
determining a unique combination of first and second data field values present in the set of previously mapped data records;
determining a code that is assigned to the data records having the unique combination of first and second data field values with an acceptable frequency; and
storing a mapping rule that maps the code to the unique combination of first and second data field values,
wherein the mapping rule is stored only for the unique combinations of the first and second data field values that occur with a frequency greater than a cutoff limit.
2. The method of claim 1 , wherein each data record comprises at least a first, a second, third, and fourth data field, the method further comprising:
determining a unique combination of first, second and fourth data field values present in the set of previously mapped data records;
determining a code that is assigned to the data records having the unique combination of first, second, and fourth data field values with an acceptable frequency; and
storing a mapping rule that maps the code to the unique combination of first, second, and fourth data field values,
wherein the mapping rule is stored only for the unique combinations of the first, second, and fourth data field values that occur with a frequency greater than a cutoff limit.
3. A method for assigning codes to prescription transaction data records, comprising:
receiving a first and a second subset of data records of the prescription transaction data records, wherein the prescription transaction records include at least a first, second, and third data field, wherein data records in the first subset have been previously assigned codes based on known mappings of the values in their respective third data fields, and wherein data records in the second subset have not been previously assigned codes;
for each data record in the first subset, identifying data records in which at least one of the first and second data fields are populated with data values;
determining unique combinations of data values of the first and second data fields present in the first subset;
determining the occurrence frequency of each unique combination of the data values of the first and second data fields in the first subset;
identifying the unique combinations of first and second data values that occur with a frequency greater than a cutoff limit X;
for each of the identified unique combination of first and second data values, determining the frequency distribution of codes assigned to data records in the first subset having the corresponding unique combination;
identifying codes in the frequency distribution for each unique combination having an occurrence frequency greater than a cutoff limit Y; and
for each identified code, storing the combination of the first and second data values and the identified code as a mapping rule for mapping data records in the second subset of data records.
4. The method of claim 3 , further comprising:
assigning codes to data records in the second subset based at least in part on matching the first and second data values of the data records in the second subset with one of the unique combinations of data values and applying a corresponding mapping rule.
5. The method of claim 3 , wherein the first subset includes historical data records.