John is in charge of the entire New Zealand Defence Force's healthcare. He has a background in emergency medicine and general practice. In a prior life, he was an engineer.
The New Zealand Defence Force is made up of the army, navy, and air force. Together they have a total of 14.2k personnel.
The main reason the Defence force has a healthcare system is to keep their personnel healthy, and fit to do their job - without putting strain on the public health system.
They also provide healthcare in aid situations - they will go to international disasters and provide help and care.
John has two main jobs within the Defence Force:
1 - Make people fit to serve - this is slightly different to what is provided by a GP in public healthcare. If you have an issue which isn't solved but doesn't affect your ability to serve, it's not a problem for them.
They are a healthcare provider in a business whose core business - is not healthcare. They are an enabler for the main function of the Defence Force - so he has to work around them.
They run essentially all of what is provided by a DHB. They provide a full primary care system for their personnel. Additionally, they have the complication of having to move it - wherever their troops go; they need to provide healthcare.
"Imagine picking up Auckland DHB, moving it to Dunedin, then three weeks later moving it to Fiji."
2 - Collect information. He needs to be able to show leadership at any time how many troops are fit to be deployed - and what the long term societal burden of service is. When a person is deployed, has exposure, that creates liability. They need to look at the longitudinal information to inform both the Defence Force and Government.
John describes healthcare models as coming in a couple of ways - usually sick-care vs wellness model. Sick-care is what most people are used to - you get sick, you go to your GP, they have a solution. Wellness models focus on keeping healthy people healthy - for example, immunisations, proper diet, contraception, public health education are all examples of well-care.
The Defence Force aims to sit half way. They provide sick-care, but they also provide well-care - as they need to ensure everybody is still fit to serve.
For example - if you're part of a healthy civilian population, providing you don't have a chronic condition, you would see your doctor 2-3 times a year. However, if you're part of the Defence Force you would make 11 points of contact a year.
John's healthcare team provides 110k patients contacts a year, usually for a simple check-up. This allows them to catch problems when they're small, and provides a huge overview of the population - every contact collects data.
The most amazing part of their system is that they provide these 110 thousand patient contacts per year, with a team of only 24 registered doctors. They do this by not using a 'doctor centric' model.
Their doctors provide the ability to manage complex cases, and they guide and govern about 850 additional healthcare staff - who do about 2/3rds of the workload. This includes nurses, and medics - personnel trained in simple healthcare, but not registered.
In civilian healthcare, the nurses operate under a limited scope of practice. However, in the Defence Force nurses and medics can examine, diagnose, and treat - all without a doctor being involved.
They use Profile for Windows. MedTech doesn't work for them - it's good for a linear model of care, and small practices - but that's not how they work.
He described the UI of Profile as clunky, but the database and backend systems as 'phenomenal' - it was one of the main reasons they chose Profile. The volume which they're using Profile (and will be in the future), has made it a worthwhile choice.
The size they could potentially have to scale to is immense. They could go from all of the Defence Force - to all of New Zealand - to all of The Coalition if needed. Seems extreme, but it's a situation they have to consider.
I asked him if they ever considered making their own software, instead of using a vendor. They have connections with many other militaries around the world - and there has been a number who have tried to self-make, and he has yet to see a successful case. There have been examples where countries put vast money into systems - in one case, more than an entire Defence budget, but were not able to crack it. He is a strong believer you need to work with a specialised health software developer.
Defence runs its own secure network, which they use. However, this makes it hard to interface with the public healthcare systems, due to a firewall, "Like the Berlin Wall during the Cold War". They run their own server system with redundancies + off-site backup. In theory, they should never go down, but in practice they do. They act as a mini EHR - an individual's records are accessible at all of their locations.
They will be one of the early adopters of the National Electronic Health Record system. They want to support it early to ensure it moves in the direction they want for the future - and are interested how far it will go.
If they deploy to another country they have two options. One, section out a part of their database, and physically take it with them. Then when they return, they'll merge it back into the main database. Technically, the most robust system, but not that practical.
John prefers to do an online deployment by making a live link back to NZ, usually via a satellite link. There are security issues to work through, but this is the most practical way.
I've asked every person I've interviewed, "if you had a magic wand, what would you change about your software". I was very impressed with John's answer, and I think it showed his leadership skills.
"There is no magic wand, no silver bullet. There is a lot of competing demands you have to manage and balance off - based off what you want to achieve, and what your workforce needs. Your workforces view is only a small slice of the pie of what you need from the full system.
If I was to ask my Doctors, it would be around the UI, and being able to rapidly get through a consultation without any delays. Everything at their fingertips - no more than 1 click away. Be able to assist them through the process of achieving the right level of documentation. And it needs to provide decision support - without them necessarily having to ask for it. It's all about time and flow. Every-time you take your attention away from the patient, that experience degrades.
If I was to look at it from the view of my non-registered, my medics. They work under a standing orders framework. As soon as they start into the documentation - it's pre-formatted for them. When they get to the diagnosis, there's a pathway and a prescription. What they need is a very templated way of doing things - to be legally compliant.
If I was to look at it from my manager's point of view - they need the ability to see whats going on. This is where tension shows between providers and managers. Providers don't understand information privacy - don't make public what shouldn't be. Two thousand years of tradition there - well the world's moved on. My managers need to have visibility over the healthcare system - both in quality of record, the flows of information, whats the demands are on the system - whats being seen, all sorts of issues. How is the logistic system going, have we transferred from this drug to that drug?
If I look at it from my commander's point of view - all they want to know is - can that person go to fight. They need to find that info without wading through pages of the system.
There is no one silver bullet - there will always be trade-offs. What we're aiming for a 90%+ compromise."
To manage their healthcare software they have two full-time staff to maintain and improve it, and one consultant who specialises in security. So they have a lot of resources and capacity to improve their system
One thing they're very interested in is applying predictive analytics to their dataset. Their data set isn't truly 'Big Data' - but it's not small. They are actively perusing how their data could be used for predictions and research.
They have a significantly more complete view of their population compared to civilian healthcare systems. They have very regular data points (almost monthly checkups), know exactly what their soldiers ate, how much exercise they did, where they went - they're in a fantastic position to test large scale data analytics.
However, John knows he cannot 100% trust the data. They're a volunteer military - the personnel are there because they want to be. When an individual gives over information, that could have adverse effects for them - at an extreme, it might mean they're unable to deploy. So they have to take all information provided with a grain of salt. The medical staff aren't just there to provide healthcare; they're also there to provide risk assessment for an individual.
Additionally, they cannot be considered Representative of the NZ population. If they produced any research based off that data, they have to be clear it cannot apply to other populations of patients.
Privacy is a common concern with any large scale analytics. John finds it fascinating as attitudes change depending on what group you talk to. People in his generation (50's) are terrified of big data. They don't understand the technology, and they don't know they already live in a world with no privacy. John thinks civilian practitioners are working in a model from this generation, which is very outdated - but they don't know the world has moved on.
However, when John tells his younger soldiers about what they're planning to do with analytics, their response is, "What, you mean you're not doing it already?"
It gets even more complicated when you consider the privacy of information generated from this dataset. Some deployments are classified - so the medical records recorded during the deployment could also be classified. Then, if they generate new information from that information, does that mean the new information is also official information? Would they have to comply with an official information request for it?
They have been talking to the Privacy Commission about what the future of health information privacy will look like, and what needs to happen.
They are currently testing 24/7 monitoring on a group of soldiers, where they are, how fast they're moving, their heart rate - everything they can record. They're doing this to refine their basic training and reduce injuries.
"In 5 years, I want the entire Defence force instrumented."
They've also been looking at what is possible with EEG instrumenting, however, they're not in a position to deal with the data volumes produced by it. I asked if they had considered genetic testing, and they're avoiding it for privacy issues.
I was very impressed at how much scope the army had to change their systems, and how far they were looking towards the future. John believed this was mainly due to the Defence Force being a goal-driven organisation. Businesses cases are put forwards, ranked by importance, and it's decided what will be pursued. When that decision is made, the whole organisation will focus on that goal, and make it happen. So they have a lot of resources to implement change, but also have a culture of change and innovation.