Elastic data privacy-compliant healthcare analytics

A system for providing data privacy-compliant healthcare analytics, comprising:

• a secure hosted environment;
• a computer that interfaces with the secure hosted environment, wherein the computer comprises at least a processor and a memory, wherein the memory is in communication with the processor, and wherein computer-executable instructions are stored on the memory and executed by the processor, said computer-executable instructions causing the processor to:
• receive, in the secure hosted environment, customer data, wherein at least a portion of the customer data comprises PHI, itself comprising personally identifiable information and/or protected health information corresponding to one or more individuals;
• create encrypted tokens for the PHI corresponding to the one or more individuals;
• receive, in the secure hosted environment, de-identified customer data, wherein the de-identified customer data includes one or more required consumer attributes;
• create health data encrypted tokens for aggregated de-identified health data from a de-identified healthcare information dataset;
• receive, in the secure hosted environment, the health data encrypted tokens for the aggregated de-identified health data;
• receive, in the secure hosted environment, a pre-built executable software package that is installed in a secure container in the secure hosted environment, wherein the pre-built executable software package comprises computer-executable code to learn and train one or more models using the encrypted tokens and the health data encrypted tokens, and wherein the pre-built executable software package includes business rules to ensure that training of the one or more models does not increase a risk of re-identification of the PHI corresponding to the one or more individuals;
• transmit outputs from the one or more models of the pre-built executable software package for review and certification by a data privacy review to ensure the risk of re-identification of the PHI corresponding to the one or more individuals remains at or below one or more thresholds of the risk of re-identification of the PHI corresponding to the one or more individuals;
• and
• delivering from the secure hosted environment, a binary of a model, statistical scoring model, or an aggregate report from the one or more models if the risk of re-identification of the PHI corresponding to the one or more individuals remains at or below the one or more thresholds as determined and certified by the data privacy review.

A non-transitory computer program product comprised of computer-executable code stored on a computer-readable medium, said computer-executable code for performing a method comprising:

• receiving, in a secure hosted environment, customer data, wherein at least a portion of the customer data comprises PHI, itself comprising personally identifiable information and/or protected health information corresponding to one or more individuals;
• creating encrypted tokens for the PHI corresponding to the one or more individuals;
• receiving, in the secure hosted environment, de-identified customer data, wherein the de-identified customer data includes one or more required consumer attributes;
• creating health data encrypted tokens for aggregated de-identified health data from a de-identified healthcare information dataset;
• receiving, in the secure hosted environment, the health data encrypted tokens for the aggregated de-identified health data;
• receiving, in the secure hosted environment, a pre-built executable software package that is installed in a secure container in the secure hosted environment, wherein the pre-built executable software package comprises computer-executable code to learn and train one or more models using the encrypted tokens and the health data encrypted tokens, and wherein the pre-built executable software package includes business rules to ensure that training of the one or more models does not increase a risk of re-identification of the PHI corresponding to the one or more individuals;
• receiving a review of outputs from the one or more models of the pre-built executable software package from a data privacy review and certifying the outputs to ensure the risk of re-identification of the PHI corresponding to the one or more individuals remains at or below one or more thresholds of the risk of re-identification of the PHI corresponding to the one or more individuals;
• and
• delivering from the secure hosted environment, a binary of a model, statistical scoring model, or an aggregate report from the one or more models if the risk of re-identification of the PHI corresponding to the one or more individuals remains at or below the one or more thresholds as determined and certified by the data privacy review.