Systems, methods and apparatus to process medical prescription order rejections

1. A computer-implemented method, executed by one or more processors, to process a medical prescription order rejection, the method comprising: submitting, using one or more processors, a first order for a medical prescription to a third-party entity for payment; and when a rejection of the first order is received from the third-party entity:

forming an input vector including data from the first order and data from the rejection of the first order; processing the input vector, with a trained machine learning model, to form a second order for the medical prescription, wherein coefficients of the machine learning model are trained by one or more processors based on training data representing a plurality of rejected orders for prescriptions and a plurality of approved orders for prescriptions by updating the coefficients of the machine learning model based on computed differences between (i) orders for medical prescriptions determined by the machine learning model, and (ii) associated approved and rejected orders;

submitting, using one or more processors, the second order for the medical prescription to the third-party entity for payment; and when the second order is approved by the third-party entity, notifying, using one or more processors, a prescription fulfillment entity that the medical prescription is available to be fulfilled.

2. The method of claim 1 , further including, prior to forming the second order for the medical prescription:

applying a plurality of rules to the first order to form a third order for the medical prescription; and

submitting the third order for the medical prescription to the third-party entity for payment,

wherein the first order is processed with the machine learning model to form the second order when the third order is rejected by the third-party entity.

3. The method of claim 2 , further including, when the third order for the medical prescription is approved, notifying the prescription fulfillment entity that the medical prescription is available to be fulfilled.

4. The method of claim 1 , further comprising, when the second order is rejected, submitting the first order for manual review.

5. The method of claim 1 , further comprising:

training the machine learning model with a first portion of the training data; and

validating the machine learning model with a second portion of the training data.

6. The method of claim 1 , wherein the plurality of approved orders includes at least one of a third order generated by applying a plurality of rules, or a fourth order generated by manual review.

7. The method of claim 1 , wherein the third-party entity includes at least one of an insurance company, a primary insurance company, a secondary insurance company, a government institution, a co-pay assistance program, a co-pay assistance organization, or a manufacturer.

8. The method of claim 1 , further comprising:

training the coefficients of the machine learning model by applying, using one or more processors, a gradient boosting algorithm to the machine learning model based on the computed differences until a metric computed over the computed differences satisfies a criteria.

9. A prescription filler system, the system comprising:

a first interface configured to receive a medical prescription;

a second interface configured to submit a first order for the medical prescription to the third-party entity for approval;

an order handling module configured to, when a rejection of the first order is received from the third-party entity, form an input vector including data from the first order and data from the rejection of the first order;

a machine learning model configured and trained to process the input vector to form a second order for the medical prescription, wherein the second interface is configured to submit the second order for the medical prescription to the third-party entity for payment, wherein coefficients of the machine learning model are trained by one or more processors based on training data representing a plurality of rejected orders for prescriptions and a plurality of approved orders for prescriptions by updating the coefficients of the machine learning model based on computed differences between (i) orders for medical prescriptions determined by the machine learning model, and (ii) associated approved and rejected orders; and

a fulfillment system configured to fill the medical prescription when the second order is approved.

10. The system of claim 9 , wherein the first interface includes the second interface.

11. The system of claim 9 , further including a rules-based handler configured to apply a plurality of rules to the first order to form a third order for the medical prescription, wherein the second interface is configured to submit the third order for the medical prescription to the third-party entity for payment, and wherein the machine learning model processes the input vector to form the second order when the third order is rejected by the third-party entity.

12. The system of claim 11 , wherein the fulfillment system is configured to fill the medical prescription when the third order is approved.

13. The system of claim 9 , wherein the third-party entity includes at least one of an insurance company, a primary insurance company, a secondary insurance company, a government institution, a co-pay assistance program, a co-pay assistance organization, or a manufacturer.

14. The system of claim 9 , further comprising a training module configured to:

train the coefficients of the machine learning model by applying a gradient boosting algorithm to the machine learning model based on the computed differences until a metric computed over the computed differences satisfies a criteria.

15. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors, cause a machine to:

submit a first order for a medical prescription to a third-party entity for payment; and

when a rejection of the first order is received from the third-party entity:

form an input vector including data from the first order and data from the rejection of the first order; an

process the input vector, with a trained machine learning model, to form a second order for the medical prescription, wherein coefficients of the machine learning model are trained by one or more processors based on training data representing a plurality of rejected orders for prescriptions and a plurality of approved orders for prescriptions by updating the coefficients of the machine learning model based on computed differences between (i) orders for medical prescriptions determined by the machine learning model, and (ii) associated approved and rejected orders;

submit the second order for the medical prescription to a third-party entity for payment; and

when the second order is approved by the third-party entity, notify a prescription fulfillment entity that the medical prescription is available to be fulfilled.

16. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions that, executed by one or more processors, cause the machine to, prior to forming the second order for the medical prescription:

apply a plurality of rules to the first order to form a third order for the medical prescription; and

submit the third order for the medical prescription to the third-party entity for payment,

wherein the first order is processed with the machine learning model to form the second order when the third order is rejected by the third-party entity.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the instructions, when executed by one or more processors, cause the machine to, when the third order for the medical prescription is approved, notify the prescription fulfillment entity that the medical prescription is available to be fulfilled.

18. The non-transitory computer-readable storage medium of claim 15 , wherein the instructions that, when executed by one or more processors, cause the machine to, when the second order is rejected, submit the first order for manual review.

19. The non-transitory computer-readable storage medium of claim 15 , wherein the plurality of approved orders includes at least one of a third order generated by applying a plurality of rules, or a fourth order generated by manual review.

20. The non-transitory computer-readable storage medium of claim 15 ,

wherein the third-party entity includes at least one of an insurance company, a primary insurance company, a secondary insurance company, a government institution, a co-pay assistance program, a co-pay assistance organization, or a manufacturer.