Real-time feedback systems for tracking behavior change

1. A system for providing real-time actionable feedback personalized to individual clinical care providers, the system comprising:

(a) a plurality of client devices, wherein each individual client device is configured to display a graphical user interface associated with a clinical care provider; and

(b) a server in communication with the plurality of client devices and a database, wherein the server comprises:

a first module configured to:

(i) transform raw clinical care data into predefined tiered data structures and store the transformed clinical care data in the database, wherein the predefined tiered data structures are suitable for being processed by a feature extraction or feature selection technique,

(ii) process the transformed clinical care data using a machine learning algorithm trained model to identify one or more performance metrics from a plurality of performance metrics to be one or more key performance metrics that impact a clinical care outcome in a selected field, and

(iii) implement an influence algorithm to identify one or more actions that influence the one or more key performance metrics, wherein the one or more actions are used to personalize a notification for a selected clinical care provider to increase a likelihood of an adjustment of the one or more actions, and wherein implementing the influence algorithm comprises at least processing the transformed clinical care data in the predefined tiered data structures using the feature extraction or feature selection technique to assess an impact of the one or more actions on the one or more key performance metrics;

a second module configured to generate a real-time measurement of the one or more key performance metrics of the selected clinical care provider by computing a variation of the one or more key performance metrics of the selected clinical care provider relative to the one or more key performance metrics of a group of clinical care providers; and

a third module configured to:

(i) upon generation of the real-time measurement of the one or more key performance metrics of the selected clinical care provider, dynamically display on the graphical user interface of a client device of the selected clinical care provider the real-time measurement of the one or more key performance metrics of the selected clinical care provider,

(ii) determine an adjustment of the one or more actions identified by the first module for the selected clinical care provider, wherein the adjustment is personalized for the selected clinical care provider based at least in part on the real-time measurement, and

(iii) determine, based at least in part on a response of the selected clinical care provider to a previous notification, one or more points in time, a frequency or a time period to deliver a next notification to the client device of the selected clinical care provider to remind the selected clinical care provider to take the adjustment of the one or more actions thereby improving an engagement of the selected clinical care provider to the adjustment.

2. The system of claim 1 , wherein the one or more key performance metrics comprise at least one of cost and quality.

3. The system of claim 1 , wherein the machine learning algorithm trained model is configured to identify an attribution of a given action.

4. The system of claim 1 , wherein the adjustment of the one or more actions is dynamically calculated based on the tracked behavior of the clinical care provider.

5. The system of claim 1 , wherein the one or more key performance metrics in the selected field are different from one or more key performance metrics in a different field.

6. The system of claim 1 , wherein the one or more key performance metrics comprise at least one of length of stay and readmission rate.

7. The system of claim 1 , wherein the adjustment is further personalized based on the response of the selected clinical care provider to the previous notification.

8. The system of claim 1 , wherein the tiered data structures comprise a top tier, a middle tier or a lower tier.

9. The system of claim 1 , wherein the influence algorithm is further improved based on feedback collected from the selected clinical care provider.

10. A computer-implemented method for providing real-time actionable feedback personalized to individual clinical care providers, the method comprising:

(a) transforming raw clinical care data into predefined tiered data structures and storing the transformed clinical care data in a database, wherein the predefined tiered data structures are suitable for being processed by a feature extraction or feature selection technique;

(b) processing the transformed clinical care data using a machine learning algorithm trained model to identify one or more performance metrics from a plurality of performance metrics to be one or more key performance metrics that impact a clinical care outcome in a selected field;

(c) generating a real-time measurement of the one or more key performance metrics of the selected clinical care provider by computing a variation of the one or more key performance metrics of the selected clinical care provider relative to the one or more key performance metrics of a group of clinical care providers;

(d) implementing an influence algorithm to identify one or more actions that influence the one or more key performance metrics of the selected clinical care provider, wherein the one or more actions are used to personalize a notification for the selected clinical care provider to increase a likelihood of an adjustment of the one or more actions, and wherein implementing the influence algorithm comprises at least processing the transformed clinical care data in the predefined tiered data structures using the feature extraction or feature selection technique to assess an impact of the one or more actions on the one or more key performance metrics;

(e) dynamically displaying on a graphical user interface of a client device of the selected clinical care provider the real-time measurement generated in (c);

(f) determining an adjustment of the one or more actions identified in (d), wherein the adjustment is personalized for the selected clinical care provider based at least in part on the real-time measurement generated in (c); and

(g) determining, based at least in part on a response of the selected clinical care provider to a previous notification, one or more points in time, a frequency or a time period to deliver a next notification to the client device of the selected clinical care provider to remind the selected clinical care provider to take the adjustment of the one or more actions thereby improving an engagement of the selected clinical care provider to the adjustment.

11. The computed-implemented method of claim 10 , wherein the adjustment is further personalized based on the response of the selected clinical care provider to the previous notification.

12. The computed-implemented method of claim 10 , further comprising improving the influence algorithm based on feedback collected from the selected clinical care provider.

13. The computed-implemented method of claim 10 , wherein the tiered data structures comprise a top tier, a middle tier or a lower tier.