Opinion: New hospital project: how options stack up

Michael Hundleby was until 2018 the senior Ministry of Health official responsible for the Dunedin hospital project. He says time is the enemy of the hospital rebuild as every month of delay will add millions to the project. 

The government has paused the new Dunedin hospital, and will either build a cheaper hospital on the new site, or refurbish the old hospital. This follows a government expert review led by Robert Rust.

The initial decision to build a brand-new hospital was made by the National government in 2017. The Bill English-led government was known for its financial prudence and accepted it was not economic to repair or refurbish the existing hospital. Instead, it decided the new hospital would be built on a central city site, or at the Wakari Hospital site. The initial cost estimate was $1.2 billion to $1.4b.

Time is the enemy of this project. Sufficient time was needed to design the hospital but every month of unnecessary delay costs millions in construction inflation. Ministers pushed for reducing the project timeline. Six months were saved on the project when I used rules allowing contracting without a lengthy tender process to obtain an urgent report on the condition of the current hospital. Following a complaint about this process, then director-general of health Ashley Bloomfield found "there were valid grounds for the ministry to have appointed Proj-X on the urgent work ... without an open tender process".

Mondelez’s announcement it was going to close its Dunedin Cadbury factory was a stroke of luck. Buying most of the site from a willing seller sped up the process. The National government authorised negotiations over the purchase to start in 2017. Choosing an alternative site with operating businesses, such as a supermarket, would have been a lengthier and more expensive process.

The Dunedin hospital budget has risen to $1.9b, with claims the new cost may "approach" $3b. But there are a few red herrings relating to the $3b cost that need to be addressed.

The first is the claim the site acquisition costs, contamination, flood risk and nearby busy roads somehow dramatically increased the construction costs. The reality is these costs are a trivial proportion of the overall cost of the project, and almost any chosen site would have these sorts of factors to deal with.

The alternative Wakari site had significant challenges, and undoubtedly the University of Otago would have been forced to shift much of its health science campus to Wakari. Shifting thousands of hospital and university staff and students away from the city would have decimated central Dunedin.

And the reality is these site costs have already been dealt with — the land has been bought and the foundations have been built.

Secondly, there is the claim there is another $400 million in costs not included in the budget — $325m decommissioning the old hospital, $45m for a pathology laboratory and the rest for 250 carparks. The government’s own Rust review is sceptical about these costs and says the refurbishment cost "should be treated with a high degree of caution" and "it may be much of this additional capital can be avoided". Why on earth would you refurbish a hospital you are abandoning? Why not just sell the site? And if it was decided demolition was required, it doesn’t need to be done straight away. Christchurch’s Princess Margaret Hospital was largely abandoned years ago but the buildings remain.

The pathology provider is a private company, so it would be a relatively straightforward exercise to contract with the private provider to provide its own lab.

And a third party could provide the 250 carparks on a commercial basis (which may require some incentive). Ngai Tahu and another private company provide carparks for Christchurch Hospital. Perhaps a tangible way for the Dunedin City Council to support the project would be for it to build and operate a hospital carpark, as was done prior to the 2011 earthquake by the Christchurch City Council.

The real issue is the price received from CPB, the builder, is, in the words of ministers, "several hundred million dollars" over the $1.9b budget. Unfortunately, CPB may currently be the only contractor in the New Zealand market capable of building the new Dunedin hospital. And, of course, if there is only one builder in the market, the builder may think it can set the price. No government should be expected to embrace a monopoly price. Unfortunately, CPB has been involved in disputes with the government over many other projects. Until more builders capable of building a new hospital of this size are in the market, it is likely the taxpayer will continue to pay a "premium" on all future large hospital projects.

The claim the project is one of the most expensive hospital projects in the southern hemisphere needs to be put in context. Once you carve off the $400m relating to refurbishment, the laboratory and carparks, the cost is likely to be in the low $2b range. The new Dunedin hospital has around 410-421 overnight beds, 53 ED beds and 15 theatres. As a comparison, in March 2024 the South Australia government announced plans for a new Adelaide women’s and children’s hospital. This hospital has 414 beds, 43 ED beds and 14 theatres with a budget of $3.4b, with the government acknowledging the cost may well increase. Australian hospitals generally are much more "flash" than their New Zealand equivalents. As an example, the Melbourne Royal Children’s Hospital has a meerkat enclosure in a vast atrium.

The first option of looking to reduce the cost of the planned hospital is the best of the two government options. There will be ways to reduce the cost without the need to significantly redesign the hospital. A redesign adds cost because of design fees and construction inflation.

Another value-management process seems to be under way. This process looks at ways to reduce cost in ways that do not unduly impact on the services, for example using cheaper tiles or cladding. And most modern hospitals build in a capacity for future growth, so it may be that the fit-out of some wards or theatres could be delayed. This has been done on recent projects such as the Christchurch hospital project where some wards were not fitted out, but subsequently added back in. The Labour government planned to do this last year, so there may be political consensus around this.

The number of floors could be reduced if the missing space could in the future be built on a nearby site, but this redesign will cause expensive delays. And because the new Dunedin hospital site physically covers most of the city block, and the spare space to the north of the outpatients building is too far away, the new space would probably need to be constructed on an adjacent site not owned by the government. And keeping active clinical services going at the old hospital building would create unacceptable clinical risk due to the distance, and there would be additional operational costs keeping both hospitals going.

The size requirement of the new Dunedin hospital could potentially be reduced if a major new hospital to serve Central Otago was planned. Queenstown’s current 15-bed hospital is woefully inadequate for the current and projected population. A new Queenstown hospital would mean many Central Otago patients would not be required to travel to the new Dunedin hospital. But a new hospital in Queenstown would be more expensive than any savings in reducing the size of the new Dunedin hospital.

The other option, refurbishment of the existing hospital, will be quickly shown, once again, to be impractical. Experts have already determined a refurbishment would be more expensive than a rebuild, and the existing structure of the hospital would mean the end result would be a second-rate hospital. Refurbishment would mean wards would constantly need to be shifted somewhere, probably to temporary decanting space, while the demolition and refurbishment took place. The site constraints and inevitable contamination would be much worse than encountered on the Cadbury site. It would also mean at least $100m-300m already spent on design, obtaining resource consent, buying the site, piling and laying the concrete foundation would effectively be flushed down the drain.

Dunedin people will be aware the business case and design process have already taken around seven years. Every month of delay will add millions to the final cost. The South Australian government estimates every month of delay to the $3.4b Adelaide women’s and children’s hospital will cost it over $10million a month. To design a refurbishment of the new Dunedin hospital, and the slow build required to refurbish an operating hospital, would add years and years to the project. And add hundreds of millions of dollars to the project.

The government is looking to extract "several hundred million dollars" from the project. Refurbishing the current hospital is not feasible. Hopefully, the project team’s value-management process will be able to come up with some significant savings on the new Dunedin hospital project that do not impact on clinical services. Maybe clinical leaders and the community would be prepared to accept deferring the fit-out of some spaces, which could be quickly included any time before the hospital opens. And perhaps the builder, CPB, could take the opportunity to review its pricing. Then we can get on with finally building what the government reviewer, Robert Rust, described as a "modern, efficient and patient-centred teaching hospital that will benefit generations of people across the southern region".