Systems and methods for medical referral analytics
A health information management system is provided which includes the ability to identify explicit referral activity reported into a referral workflow system, infer referral activity not reported into the referral workflow system utilizing intent-based clustering of medical information, and generate reporting metrics from the inferred and explicit referral activity. Additionally, a referral suggestion may be generated for an identified condition. Physicians are selected within a geographic area who are properly specialized for the condition. This group is then filtered based upon patient preferences and default preferences in order to generate physician referrals.
1. In a health information management system, a method executed on a computer processor for referral analysis, the method comprising:
aggregating data from multiple health networks into a consolidated data lake;
meta tagging data from the data lake for a unique health network and use case at each processing units of a plurality of parallel processing units;
generating separate data harbors for each health network from the meta tagged data;
identifying, by the computer processor using a data harbor associated with a given health network, explicit referral activity reported into a referral workflow system;
inferring referral activity, by the computer processor, not reported into the referral workflow system by:
collecting medical records from the data harbors;
identifying terms in a first set of the medical records that indicate a medical event through conceptual models using machine learning;
determining when the medical event is a severe condition based upon rules;
determining a suggested event based upon the conceptual models for the severe condition;
identifying if terms in a second set of the medical records that are after the first set of the medical records indicate a referral event matching the suggested event; otherwise
determining that no referral event occurred;
generating reporting metrics, by the computer processor, from the inferred and explicit referral activity;
generating a referral suggestion, by the computer processor, by identifying a condition, activity, procedure, or diagnosis for the referral, selecting physicians within a geographic area who are properly specialized for the condition, activity, procedure, or diagnosis, and filtering the physicians by at least one of patient preferences, default preferences, referring physician history, quality of treatment, and network preferences; and
displaying the referral suggestion to a patient via a web portal on a computing device.
2. The method of claim 1 wherein the intent-based clustering of medical information further comprises:
reconciling the medical information; and
clustering the reconciled medical information, wherein the clustering includes applying at least one clustering rule to the reconciled medication information.
3. The method of claim 2 wherein the reconciling further comprises applying at least one similarity rule.
4. The method of claim 3 wherein the at least one similarity rule includes comparing patient data attributes.
5. The method of claim 4 further comprising computing a distance between the patient data attributes and clustered reconciled medical information.
6. The method of claim 5 wherein if the distance is less than a threshold, then the clustered reconciled medical information is included as a referral.
7. The method of claim 1 further comprising:
aggregating patient health records from across multiple health networks into a data lake;
designating a question of value for a health network;
iteratively querying the data lake to determine data relevant to the question;
including the relevant data in an analysis to determine performance gains.
8. A health information management system, including non-volatile machine readable data being executed on a processor, comprising:
a data manager comprising a processor for aggregating data from multiple health networks into a consolidated data lake;
separate parallel processing units comprising a processor for meta tagging data from the data lake for a unique health network and use case at each processing unit a plurality of data bases comprising a processor for generating separate data harbors for each health network from the meta tagged data;
a referral collector comprising a processor executing the non-volatile machine readable data, configured to identify explicit referral activity reported into a referral workflow system using a data harbor associated with a given health network;
an inference engine comprising a computer system executing the non-volatile machine readable data configured to infer referral activity not reported into the referral workflow system by:
collecting medical records from the data harbors;
identifying terms in a first set of the medical records that indicate a medical event through conceptual models using machine learning;
determining when the medical event is a severe condition based upon rules;
determining a suggested event based upon the conceptual models for the severe condition;
identifying if terms in a second set of the medical records that are after the first set of the medical records indicate a referral event matching the suggested event; otherwise
determining that no referral event occurred;
an analyzer comprising a computer system executing the non-volatile machine readable data configured to generate reporting metrics from the inferred and explicit referral activity;
the analyzer further configured to generate a referral suggestion by identifying a condition, activity, procedure, or diagnosis for the referral, selecting physicians within a geographic area who are properly specialized for the condition, activity, procedure, or diagnosis, and filtering the physicians by at least one of patient preferences, default preferences, referring physician history, quality of treatment, and network preferences; and
an interface comprising a web portal on a computing device for displaying the referral suggestion to a patient.
9. The health information management system of claim 8 wherein the inference engine further comprises:
a reconciliation engine configured to reconcile the medical information; and
an intent-based presentation engine configured to cluster the reconciled medical information, wherein the clustering includes applying at least one clustering rule to the reconciled medication information.
10. The health information management system of claim 9 wherein the reconciling further comprises applying at least one similarity rule.
11. The health information management system of claim 10 wherein the at least one similarity rule includes comparing patient data attributes.
12. The health information management system of claim 11 wherein the intent-based presentation engine is further configured to compute a distance between the patient data attributes and clustered reconciled medical information.
13. The health information management system of claim 12 wherein if the distance is less than a threshold, then the clustered reconciled medical information is included in a presentation cluster prepared for the user.
14. The health information management system of claim 8 further comprising a big data analyzer configured to perform the steps of:
aggregating patient health records from across multiple health networks into a data lake;
designating a question of value for a health network;
iteratively querying the data lake to determine data relevant to the question;
including the relevant data in an analysis to determine performance gains.