Dashboard week has begun, and we started by utilizing API hospital data from the Australian Institute of Health and Welfare Government website (found at MyHospitals Data API (aihw.gov.au)). To begin, we needed to download the data from the API URL and JSON parse it in Alteryx. This was followed by cleaning the data and filtering to the exact datasets we needed. A batch macro was required so that once we figured out which datasets we wanted to use, we were able to retrieve all the different years the dataset had that were available by continuously running the workflow until all the years were added to the data table. Below is my Alteryx workflow.

While there are 4 different datasets outputting from the workflow, I ultimately ended up only using 3 (removed MYH0009) as it was taking around 11 minutes each time I wanted to either run the workflow or execute a query in Tableau. Below is what my dashboard ended up looking like (this was after presenting it and receiving a couple pointers from Beth which I ended up adding to my dashboard).

 

Going from the top to the bottom of the dashboard, clicking on the Government image takes you to the webpage for the API data. The measure values legend is for the map, and the measures can be switched between the sum and averages of ‘number of admissions to hospital’ and ‘number of overnight hospital stays’. There’s also a dropdown that allows you to switch between the top and bottom hospitals in the bar chart, and next to it is a parameter that allows you to change what number of top or bottom hospitals you can filter down, or up to. Finally, there is also a parameter that allows you to change which year of data you can look at for the visuals.

Next are a couple of BANs with information on the total number of hospitals per state, the total amount for the measure chosen for each state, which changes depending on the year chosen on the parameter, and the percentage of overnight stays for each state and year chosen, which also shows the percentage difference from the previous year.

Moving down, we have a bar chart that shows the measure chosen for each state in a visual way that allows you to see how they rank against each other, and then below is a line chart that shows the chosen measure over time by years, and when you click on the state bar chart (or map and hospital bar chart), the line chart will filter to a certain state and/or a certain hospital. Additionally, when you hover over these visualizations, you will be able to see the most common reporting measures (reasons for hospital admissions or overnight stays) for each state or year.

Finally, the state map allows you to see the top/bottom hospitals per state when clicked on, based on whichever measure is chosen. You can also see the most common reporting measures in the tooltip when you hover over both the state map or the hospital bars.

You can find my dashboard on Tableau Public:

https://public.tableau.com/views/HospitalAPIDashboard/Dashboard?:language=en-US&publish=yes&:sid=&:display_count=n&:origin=viz_share_link

 

Nicholas Seah
Author: Nicholas Seah