High Level Design
Overview
In order to improve the operations and quality of care, hospitals take advantage of Sisense to monitor key metric across all departments. In the case below, a hospital is monitoring admissions by a department and comparing the number of doctors and average wait time. Indications like this can help determine the necessary resources needed to run a department. Monitoring the patient satisfaction provides a way to measure the performance of the doctors working in the hospital as well as the overall quality of each division.Goals
Increase hospital operational efficiency while improving patient care experience and quality.
Objectives
- Maintain a pleasant patient experience
- Monitor an adequate doctor staff to patient work load
- Manage hospital costs and treatment
KPI Architecture
| KPI | Measures | Calculation | Table |
| Patient Satisfaction | Patient Satisfaction | Dupcount ([Patient Satisfaction]) | Survey |
| Doctors Treatment Plan | Contribution ([dupcount([Doc Explanation])]) | Survey | |
| Confidence in Treatment | Dupcount ([Doc Confidence]) | Survey | |
| Operational Workload | Patients per Doctor | Avg ([Months in date],[Total Patients])/avg([Months in date],[# of unique Employee ID]) | Admissions |
| Number of Doctors | Count ([Employee ID]) | Employees Work | |
| Operation Costs | Average Cost per Patient | [Total Cost]/sum ([Inpatient Admission])-[Total Patient discharge]+[Total Outpatient Admission] | Admissions, Treatments |
| Operation Costs | Sum ([Cost]) | Treatments | |
| Patient Volume | Number of Patients | Sum ([Inpatient Admission])-[Total Patient discharge]+[Total Outpatient Admission] | Admissions |
| Admitted Patients | Sum ([Inpatient Admission])-[Total Patient discharge] | Admissions | |
| # of Out Patients | [Total Outpatient Admission] | Admissions | |
| Patient Experience | Avg Wait Time | Avg ([Wait time]) | Treatments |
Data Modeling (ElastiCube Design)
Data Security
In the following example, a row has been created for every department-user combination based on their user regulations. The first field ‘Division’ connects to all the fact tables. The second field ‘UserIdentifier’ corresponds to the username that should have access to those departments.
Notice how Divisions and UserIdentifiers can repeat, however the combination is still unique? The method we are using will allow us to create a relationship between the user and the dimension where we can control what divisions a user can see in the data source and the admin only has to apply their name to set the data security rule.
Note: There is a row for every category assigned to Admin. This addition will allow you to create a user that can see all possible selections.
The next step is to apply the security rule. Go to Admin >Data Sources and select the menu option for the data source of your choice and click on Data Security. Once inside, select the UserIdentifier field and set the appropriate values or data security.
Note: using this method will create a many to many relationship when a user accesses a dashboard with no selection made. Therefore, once implemented a user must always be selected when data security is not applied.
To test out the data security try logging in as a test user with data security and notice how the data changes.
- User: John “J.D.” Dorian
- User 2: Christopher Turk
Plugins & Scripts Used
Data Connectors
