Canterbury District Health Board drives self-service analytics


The Burwood Health Campus, located in Christchurch, New Zealand.

The Canterbury District Health Board (DHB) is the main planner and funder of health services in the Canterbury region of New Zealand. One of Canterbury DHB’s key strategies is to leverage data to drive an integrated health system. The DHB needed to strengthen its data analytics and visualization capabilities. With almost 20 years’ worth of data stored in a data warehouse, static reports along with ad-hoc querying made it difficult for the business to see and understand their data. The Decision Support team sought a solution that would let a wide group of staff easily interact with data, while empowering the wider workforce to make more data-driven decisions. In three years, Canterbury DHB has transformed the way that employees view and interact with data, including employees from other District Health Boards within New Zealand. Enabling health professionals to benefit from self-service data analytics was a driving factor in adopting Tableau as its preferred visual analytics software. Tableau is now one of the key tools empowering the Canterbury DHB to drive important decisions concerning the health and wellbeing of New Zealanders.


Turning millions of patient activity events into interactive content

Canterbury DHB holds nearly 20 years of patient activity data in a data warehouse, each associated with its own set of metrics. Canterbury DHB is responsible for planning, funding, and provisioning healthcare services to a population of over 540,000 residents. Over the course of a year, this includes almost 120,000 hospital admissions, over one million outpatient appointments, and approximately 100,000 emergency department attendances. In addition, Canterbury DHB also collects data from pharmacy and radiology services, operating theatres, and laboratory tests. Over almost 20 years, that amounts to almost 500 million events, managed by the Decision Support team. The health board also collects data from other South Island DHBs. All of this data is stored in an enterprise data warehouse. With this data, the team responds to information requests from across the organization and generates reports. As requests for data grew, the team realized that fixed reports were no longer serving the needs of the health board. John Wilkinson, Decision Support Manager at Canterbury DHB explains, “The data we are managing gets very big when you think about every procedure, every drug issued to a patient, every diagnostic test, every inpatient and outpatient event, and much more. It was all being accessed in an inefficient manner. We’d been providing fixed-format reports for many, many years,” he says.

The data we are managing gets very big when you think about every procedure, every drug issued to a patient, every diagnostic test, every inpatient and outpatient event, and much more.

If the requestor had follow-up questions, or wanted to view the data in a different way, the Decision Support team would have to rerun the report—causing delays. “We constantly had to re-run those reports with different selections in order to get a slightly different view of the data.” And the Decision Support team had no way to track how many people were accessing each report. Canterbury DHB wanted to encourage informed decision making across the business. Part of this goal was enabling employees to readily interact with data in a self-service, interactive format. “The ability to consolidate and filter the data required for each stakeholder was a main focus for us,” said John. The decision to purchase an analytics solution went beyond time savings. Leadership wanted to create a more data-driven culture throughout the organization. “We were interested in finding a solution which would ensure a good uptake of self-service analytics and reinforce distributed, informed decision-making across the workforce,” said John.

Introducing data analytics to the DHB workforce

A team member had heard about Tableau from respected people in the industry. He decided to download the 30-day Tableau Desktop trial at home and had positive results. “After hearing about the trial and what my staff member was able to achieve using small sets of data, I was intrigued and started wondering if this was something that would work on an organizational scale,” he says. Following a quick office demo using a sample data set, John was impressed. He purchased two Tableau Desktop licenses and the team began working with data in new and interesting ways. He challenged his team to use Tableau and see what they could produce; the engagement levels of his staff rapidly increased, prompting John to purchase another two licenses for his team. Over the course of time, John has provided all of the Decision Support team with Tableau Desktop, along with other like-minded, analytical staff members in other departments. Montage—a Tableau partner since 2008—supported Canterbury DHB at the beginning of their Tableau deployment. Montage conducted initial mentoring sessions to better understand their processes. They also utilized Tableau workbooks that had been successful for other DHBs. After using Tableau Desktop for a few months, Canterbury DHB started to share visualizations across the business using Tableau packaged workbooks. However, the team wanted to encourage collaboration and promote a single source of truth. They quickly realized they would need a better strategy for sharing interactive dashboards. “Constantly pulling data and updating the workbooks was going to be an expensive and time-consuming activity so we decided to have a look at Tableau Server, which enabled us to automate a lot of the work we had been doing,” John said.

More than a dashboard

The team’s ability to analyze and share data has delivered far-reaching benefits. John and his team can now effortlessly provide automated dashboards through Tableau Server, embedded in a SharePoint portal. As a result, staff access business intelligence data from a single source of truth to inform their decision making. One of the principles of the Decision Support team is to enable staff to “manage by fact, not anecdote.” With Tableau Server, John’s team can easily refresh the data in a visualization, instead of manually updating reports when the data changes. And employees can access views through a web browser, leading to faster turnaround times. John said, “The fact we can automate through Tableau Server means that we can turn data around quickly on a regular basis – some as frequently as every 15 minutes.” And in Tableau Server, the team can track the popularity of each view, revealing the data that is most relevant to business users. In addition to satisfying the information needs of Canterbury DHB, the Decision Support team is beginning to manage data on behalf of its sister DHB on the West Coast of the South Island. Canterbury manages the data, ensuring data quality. The team then creates dashboards—shared through Tableau Reader—to South Island district health boards at no extra effort. “A real benefit we’ve identified is that we can manage the data sets for multiple organisations through one application…This type of functionality lets us take a lead role in assisting other South Island DHBs.”

The fact we can automate through Tableau Server means that we can turn data around quickly on a regular basis – some as frequently as every 15 minutes.

Building an army of data heroes

Within three years of adopting Tableau as a primary analytics tool, Canterbury DHB has seen more people using Tableau, along with an increasing demand for data. Today, Canterbury DHB manages a portfolio of 45 Tableau workbooks, with approximately 100 individual views—and growing. Today, the Canterbury DHB uses Tableau to explore specific cohorts of patient activity, providing rapid insights into previously under-utilised data. Through automation and self-service, Canterbury DHB has been able to foster a data-driven, informed approach in developing plans and processes. John and his team work with data to help influence decisions on a daily basis. The team can now visualize patient-experience data to identify outliers—and predict areas where they could better service customer health needs. As more employees engage with Tableau, Canterbury DHB has been able to grow its Division Support team with more qualified, engaged employees. John says, “More satisfied, highly engaged employees means we’re able to attract a high caliber of applicant. This means the work my department is responsible for producing is of higher quality, so we get more visibility throughout the organization and it’s easier for me to manage as a result.” With Tableau Desktop, Server, and Reader, the Decision Support team saves time in their reporting cycle—while serving the needs of the business. As more people adopt the software, the organization continues to move towards a more data-driven approach to healthcare.