Method, apparatus, and computer program product for using machine learning to generate an offset amount
A method, apparatus and computer program product are provided for generating offset amounts for inquiry data objects received over a network. Machine learning algorithms may be used to determine the offset amount, based on a base quantitative amount, a product identifier, and a longevity indicator. Historical data may be modelled and analyzed to determine relationships between quantitative amounts, longevity indicators, and adherence data. Adherence may therefore be improved by generating an offset amount dependent on the machine learning algorithms and model. By utilizing a model and machine learning algorithms, the model may account for new products, different pricing structures, and/or the like, and may enable the efficient generation of offset amounts for hundreds or thousands of products in an efficient manner, and for different associated longevities.
1 . A computer-implemented method for dynamically generating an offset amount based on longevity, historical data and associated adherence, the computer-implemented method comprising:
generating a model based on a plurality of inquiry data objects from a plurality of client devices in a network, each inquiry data object comprising a respective product identifier, a respective longevity indicator comprising a respective number of refills, a respective quantitative amount, and respective adherence data, wherein the quantitative amounts and the respective number of refills are predictors of the respective adherence data;
receiving from a client device, an inquiry data object comprising at least a product identifier and a longevity indicator comprising a number of refills;
generating and transmitting a prescription benefit inquiry to a third party computer;
receiving, from the third party computer, a prescription benefit inquiry response comprising a base quantitative amount, wherein the base quantitative amount is not dependent on the number of refills;
performing an offset calculation protocol to generate the offset amount by which to offset the base quantitative amount, wherein the offset calculation protocol utilizes at least the product identifier and the number of refills, wherein performing the offset calculation protocol comprises applying the inquiry data objects to the model to calculate the offset amount such that a predicted associated adherence data for at least one inquiry data object is improved relative to a predicted associated adherence data with no offset amount, wherein an offset amount for one inquiry data object is different from another offset amount calculated for another inquiry data object having common product identifiers, common base quantitative amounts, and different numbers of refills; and
transmitting the base quantitative amount and the offset amount to the client device, wherein performing the offset calculation protocol, utilizing at least the product identifier and the number of refills, and transmitting the base quantitative amount and the offset amount to the client device results in reduced prescription abandonment and corresponding reduced computer resource consumption otherwise expended in facilitating prescription reversals caused by prescription abandonment, in comparison to prescription abandonment and computer resource consumption expended based on computer-implemented methods that do not perform the offset calculation protocol.
2 . The computer-implemented method of claim 1 , further comprising:
adjusting the offset amount based on an offset limitation.
3 . An apparatus for dynamically generating an offset amount based on longevity, historical data and associated adherence, the apparatus comprising at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least:
generate a model based on a plurality of inquiry data objects from a plurality of client devices in a network, each inquiry data object comprising a respective product identifier, a respective longevity indicator comprising a respective number of refills, a respective quantitative amount, and respective adherence data, wherein the quantitative amounts and the respective number of refills are predictors of the respective adherence data;
receive from a client device, an inquiry data object comprising at least a product identifier and a longevity indicator comprising a number of refills;
generate and transmit a prescription benefit inquiry to a third party computer;
receive, from the third party computer, a prescription benefit inquiry response comprising a base quantitative amount, wherein the base quantitative amount is not dependent on the number of refills;
perform an offset calculation protocol to generate the offset amount by which to offset the base quantitative amount, wherein the offset calculation protocol utilizes at least the product identifier and the number of refills, wherein performing the offset calculation protocol comprises applying the inquiry data objects to the model to calculate the offset amount such that a predicted associated adherence data for at least one inquiry data object is improved relative to a predicted associated adherence data with no offset amount, wherein an offset amount for one inquiry data object is different from another offset amount calculated for another inquiry data object having common product identifiers, common base quantitative amounts, and different numbers of refills; and
transmit the base quantitative amount and the offset amount to the client device, wherein performing the offset calculation protocol, utilizing at least the product identifier and the number of refills, and transmitting the base quantitative amount and the offset amount to the client device results in reduced prescription abandonment and corresponding reduced computer resource consumption otherwise expended in facilitating prescription reversals caused by prescription abandonment, in comparison to prescription abandonment and computer resource consumption expended based on apparatuses that do not perform the offset calculation protocol.
4 . The apparatus of claim 3 , wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to at least:
adjust the offset amount based on an offset limitation.
5 . A computer program product for dynamically generating an offset amount based on longevity, historical data and associated adherence, the computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions comprising program code instructions to:
generate a model based on a plurality of inquiry data objects from a plurality of client devices in a network, each inquiry data object comprising a respective product identifier, a respective longevity indicator comprising a respective number of refills, a respective quantitative amount, and respective adherence data, wherein the quantitative amounts and the respective number of refills are predictors of the respective adherence data;
receive from a client device, an inquiry data object comprising at least a product identifier and a longevity indicator comprising a number of refills;
generate and transmit a prescription benefit inquiry to a third party computer;
receive, from the third party computer, a prescription benefit inquiry response comprising a base quantitative amount, wherein the base quantitative amount is not dependent on the number of refills;
perform an offset calculation protocol to generate the offset amount by which to offset the base quantitative amount, wherein the offset calculation protocol utilizes at least the product identifier and the number of refills, wherein performing the offset calculation protocol comprises applying the inquiry data objects to the model to calculate the offset amount such that a predicted associated adherence data for at least one inquiry data object is improved relative to a predicted associated adherence data with no offset amount, wherein an offset amount for one inquiry data object is different from another offset amount calculated for another inquiry data object having common product identifiers, common base quantitative amounts, and different numbers of refills; and
transmit the base quantitative amount and the offset amount to the client device, wherein performing the offset calculation protocol, utilizing at least the product identifier and the number of refills, and transmitting the base quantitative amount and the offset amount to the client device results in reduced prescription abandonment and corresponding reduced computer resource consumption otherwise expended in facilitating prescription reversals caused by prescription abandonment, in comparison to prescription abandonment and computer resource consumption expended based on computer program products that do not perform the offset calculation protocol.
6 . The computer program product of claim 5 , wherein determining the base quantitative amount comprises generating and transmitting a prescription benefit inquiry to a third party computer and receiving a prescription benefit inquiry response comprising the base quantitative amount.
7 . The computer program product of claim 5 , wherein the base quantitative amount is not dependent on the number of refills.
8 . The computer program product of claim 5 , wherein the computer-executable program code instructions further comprise program code instructions to:
adjust the offset amount based on an offset limitation.