__STYLES__
The scope of this project was to develop a high-level KPI report for the executive team of the Massachusetts General Hospital.
The data provided by Maven Analytics goes from January 2011 to February 2022. The dataset is composed of information about the patients, the encounters they had in the hospital, the procedures that took place during the encounters, the payers covering the costs and the organizations. We have explanatory data about the patients (name, gender, location, ethnicity), the duration and the costs generated for both the encounters and the procedures.
To suit the need of KPI report, I decided to go for a Canvas letter as layout.
The idea was to ensure that the audience could easily understand the report and how to interpret the data. So first, I put the main KPIs as cards to give a first overview of the data, and then I put the main filters. I made a selection of visualizations that met the requirements and I put on one side a guiding text on the main outputs of the visualization and on the other side the visualization to guide the audience.
For this question, I configured the visualization so that you can choose between seeing the total number of admissions and readmissions over time and seeing how many times each patient has been admitted, thanks to buttons that change the visualizations accordingly.
This provides both a global and a more detailed view of the issue.
Here, as the figures are in the same scale, often similar to one another, I preferred to use a matrix for the representation to make the data easier to read. The unit is in hours, for greater clarity. With the buttons, it's also easy to see how long patients have stayed in the hospital, according to their ethnicity and gender, to push the analysis a little further with more information at hand.
For this visualization I've included both the average cost per visit and the average of procedures base cost. It's interesting to have these two sets of cost information for comparison. This gives a more comprehensive idea of the costs of each visit and the procedures that took place during the visits.
I used a matrix of institutions and the number of procedures they cover. To have more information on the cost, I decided to add a conditional formatting with the percentage covered on the total cost of the procedure by each insurance. Red being those that cover a lower proportion of costs and green a higher proportion.