Introduction

Today marks the beginning of dashboard week. A week where we have to complete an entire dashboard per day, each with unique associated tasks. Todays was to contact an API which contains a plethora of information associated within hospitals around Australia (Link to API: https://www.aihw.gov.au/reports-data/myhospitals/content/api).

Alteryx

The first section of tasks to tackle was downloading and engineering the data in Alteryx. From the API, I needed to download all the associated data for a given measure (number of surgeries for example). This was done by contacting the specific API endpoint for that reported measure, and they transforming the data as required.

I then ran this data into a batch macro which would return the time period the specific data was collected in. These endpoints on contained data from the 2012 and 2013 financial year.

The final changes I made in Alteryx were associated with the descriptions present in the measure columns. For example, it I wanted to visualise the number of surgeries, I would get errors due to the string values (such as ‘between 10 and 30’). Thus, where the string descriptions were present, I’ve replaced them based on what the most appropriate reasoning.

Tableau

Now that I could bring this data into Tableau, I wanted to look at the performance of these hospitals, particularly when it came to the dealing of different cancer types (It is also important to recognize that behind every number in this dataset that there is a traumatic experience had by someone. So looking at this data with the upmost respect is required.) . This dashboard first describes the data at a state level, which can then be broken down to the primary health network and hospital levels. What I found interesting about this data is that the number of surgeries does not correlate well with the median waiting time. For example, when looking at NSW and NT, we can see that they are polar opposites when it comes to the number of surgeries. However, NT has the longest median waiting times, whilst NSW runs in the middle of the pack. This could be indicative of the number of surgeons available in those areas, as more surgeons can take on more surgeries, leading to lower wait times.

You can check out the dashboard on my tableau public.

https://public.tableau.com/app/profile/jay.mcintosh/viz/AustralianSurgeryAnalysis-2013FinancialYear/SurgeryAnaysis

Source: https://www.aihw.gov.au/reports-data/myhospitals/content/api