Introduction

For our first day of dashboard week we were given a link to The Australian Institute of Health and Welfare’s API page and given free range to create a dashboard from anything we could find there. Starting dashboard week with API’s means it’s going to be a tough week but I got stuck in and by around midday I had used a batch macro in Alteryx to create my full dataset exploring elective surgeries and their wait times.  I pulled form four different end points, one for the total amount of surgeries, the median wait time, the percent of patients that had to wait over 365 days for their surgery, and finally the percent of patients that received their surgery within the recommended time.  My workflow was a little bit chaotic but it worked and the focus was on the dashboard so that came next.

(AIHW’s API Page)

 

The Dashboard

It took a while for my to figure out my plan of action and after exploring the data I was struggling to find any insights.  I decided I should start working on some charts and putting together the dashboard, maybe I’ll find some ideas along the way.  I brought up Excalidraw and created a few rough drafts.  I knew I’d want to see the top and bottom Hospitals and Surgeries and perhaps wait time over the years.  So after plotting that I knew I had something and got to work with pulling it all together.

 

My dashboard starts with some BANS that summarize the four main measures as listed above. I followed those with some sparklines and a percentage of how they have changed from their first instance.  Following this I created some bar charts, both have charts in their tool tips too showing the top 5 surgeries in either wait time or a measure parameter I created to demonstrate the 3 other variables.  The chart on the left shows that surgeries have increased a lot since 2016 however wait time and patients waiting over 365 days has increased drastically.  This is more than likely due to COVID however as we see the spike around then.  The chart on the right just shows which hospitals are performing best and worst for wait time.

 

 

Following this I have a Map that shows the measure parameter over states and when you click on a state a bar chart appears that allows the user to select either the top or bottom 5 hospitals or surgeries in that state.  A lot of parameters and dynamic zones were used here but it adds a level of interactivity and exploration to the dashboard that I thought made it worth it.  It is here that I found that cosmetic surgery is almost always in the worst five surgeries for wait time in all states.

 

 

Below the Map I have grouped the Surgery Types into categories and plotted the top 5 over time.  This allows the user to see how the most common surgeries compare and the chart is also filtered by the map allowing a state view of these surgery categories.  Interestingly, I found again that cosmetic surgery is one of the most common surgeries which would explain the jump in wait time recently.  QLD also has cosmetic surgery as their most common consistently however it does not see the rise that other categories did throughout COVID.  That may be as it is not essential and people waited until after hospitals were so busy.

 

 

Conclusion

While I did have a lot of fun making this dashboard, I wish I got into the dashboard earlier as a lot of the day was spent on the data preparation.  I think with more time I could have added some deeper layers, looking more into the hospitals and their most popular surgeries.  I would have also spent a bit longer on the formatting as I think it could use some more customization.  All in all though, I think it is a great dashboard that provides some interesting insights and interactivity.  If you want to check it out yourself you can at the link below!

 

TABLEAU LINK: https://public.tableau.com/views/HospitalDashboard_17198334645930/Dashboard1?:language=en-US&publish=yes&:sid=&:display_count=n&:origin=viz_share_link

API LINK: https://myhospitalsapi.aihw.gov.au/index.html

Mikael Nuutinen
Author: Mikael Nuutinen