JOIN HINZ   |   Print Page   |   Contact Us   |   Report Abuse   |   Sign In Features

Data and the pandemic – Waitematā’s story

Monday, 10 August 2020  
Share |

Return to home page


Picture: Sharon Puddle looks at a regional dashboard with i3 clinical data analysts Jane Yang and Monique Greene editor Rebecca McBeth


As Covid-19 hit New Zealand’s shores, Waitematā DHB’s data team joined the Northern Regional response. From that moment on, “every day was like a week”, as the team worked around the clock to prepare for a massive outbreak of the virus.


Virus time

When Waitematā DHB’s head of Analytics Delwyn Armstrong was first pulled into the regional response team for Covid-19 in early March this year, she asked the intelligence lead for the northern region, ‘how long do you think we've got’, 

“At that point we really didn't know what we were facing, but we were told we had around three weeks, so we thought: ‘right, what can we do that's useful in three weeks’?”

At that point they were planning for an inundation of cases and for the hospitals to be overwhelmed. 

Analytics teams from around the region, which includes Auckland, Counties Manukau and Northland DHBs, “leapt into a lot of activity”, first focusing on how to get a regional view of hospital capacity and occupancy and work started on designing a regional data store.

“We talked about meeting again tomorrow, but the infectious diseases specialists said ‘we're on virus time, tomorrow is too late’,” Armstrong says.

The group met again later that day and from that moment, “every day was more like a week”, she explains. 

“There was no weekend, everyone was so focused on what we could do before this hits us.”

Taking a regional view

Within three weeks, they had a regional data store with all Covid-19 testing information, and soon after they had visibility of all hospital capacity and occupancy in real time across the region: something never achieved before. 

Due to the success of New Zealand’s lockdown, the hospital view initially had a lot of “empty white space” rather than the overflow expected, but as testing ramped up the dashboard was used to display a regional view of results.

The regional dashboard, built on Qlik Sense, is now being used to monitor people in quarantine or managed isolation. 

Armstrong says there had been previous discussions about setting up a regional real time data store, but “we couldn’t get that off the ground as there wasn't the will to share information like this, but suddenly all the barriers dropped”.

The real-time hospital view could have uses beyond Covid-19, such as viewing occupancy of neonatal beds across the region, as currently this involves ringing around all of the units to find a space.

Notifications and pathways

Notifications for communicable diseases such as Covid-19 have traditionally been sent via fax from the three Auckland metro DHBs to the Auckland Public Health Service. However, as the testing regime ramped up, an alternative to paper was clearly needed.

Waitematā digital lead and anaesthetist Lara Hopley led the development of an electronic notification system for Covid-19, used by the Auckland metro DHBs.

The region’s e-referrals system was repurposed to allow GPs to notify of suspicion of Covid-19, with information fed directly to the Public Health Servie and the regional data store. The system has had more than 190,000 notifications so far.

The team is now looking at expanding the service to cover all communicable diseases.

In the first week of work, Hopley also led development of an electronic Covid-19 monitoring electronic care pathway, which meant that hospitals knew who they were monitoring for infection.

The ever increasing states of suspicion of Covid-19 that could be recorded also meant the team had to develop a Covid-19 Status Logic Engine: an algorithm to understand the Covid-19 status of every patient, including ‘under investigation’ and later on, ‘recovered’. 

“It's really been fascinating because what I realised was everything that we deal with in data over-simplifies medicine, as all the hard work has been done to boil things down to a binary state, such as a person has diabetes or doesn't,” Armstrong explains.

“But actually, underneath it is a complex disease process and people have gone through decades to try and boil it down to a cutoff point for yes or no. None of that existed with Covid-19 because it was a brand-new disease, so knowledge about the disease was emerging day by day. 

“We were tracking it in the data, working with clinicians to decide what our rules would be. It was fascinating to be part of a process that you're not normally involved in,” she tells

Working out who to go to war with

Armstrong says a key lesson learned from the experience was to “build on your foundations: use the systems you know and build on them and don't start anything from scratch”.

It was also key to use people who can make decisions, but also get involved in the detail. 

“What we saw was that if you separated the decision making from the detail people, it wouldn't work because you didn't have time for a ‘planning phase’, then a ‘doing phase’, both had to start at the same time,” she says.

She describes chief information officer Stuart Bloomfield and i3 director Penny Andrew as key enablers of the pace of work required as they could make decisions, but were also prepared to get involved in the details.

“We are also lucky to have highly skilled, knowledgeable data analysts, data managers and data savvy clinicians across the region,” she says. 

“The will to get things done was unbelievable over this time: you worked out who you wanted to go to war with.”

Quantifying Covid-19

Head of digital transformation at i3, Sharon Puddle, says the DHB has been working with the Health Quality and Safety Commission over the past two months to look at the impact of Covid-19 and specific metrics that they might want to measure in comparison with other DHBs. 

“You can quantify the effect of Covid-19 by predicting what we would have had without Covid-19, and what it was with Covid-19, and then look at that for vulnerable groups like Maori, Pacific and children,” she explains. 

This data has shown that Maori and Pasifika people were getting tested at appropriate levels and the detection rate was lower than average, but that Pasifika people were later to return to hospitals than other groups. 

Dashboards were also created to look at emerging vulnerable groups like people in aged residential care and healthcare workers

Data is key to the DHBs planning to enable it to ‘catch-up’ on services and care that was unable to be delivered during lockdown.

“Having these analytics tools means that we have a lot of that data and can turn it over really quickly and present it to the decision makers. Then they can put in different scenarios to look at how we can shape our services,” Puddle says.

“We’re looking to use various data sources to target the people who've been waiting the longest and who have the greatest need, on the waitlist.”

One unexpected, but positive use of the data, was to identify people in hospital who had a birthday coming up. As inpatients were not allowed visitors during lockdown, the patient experience team were able to visit them and set up calls with family members, or provide simple things like magazines for them to read.

“There's little things that you don't think about when you bring the data together, but it was really nice to hear how it can be used,” Puddle says.


Hear more from Delwyn Armstrong during a free eHealthNews Live Webinar at 12.30pm on October 28, 'business intelligence driving healthcare transformation'. 


If you would like to provide feedback on this news story, please contact the editor Rebecca McBeth.


Read more features:

Cloud enables Covid scale-up
Creating resilience through cloud technologies

Return to home page

HiNZ, PO Box 300125, Albany, Auckland 0752, New Zealand.

Membership Management Software Powered by YourMembership  ::  Legal