22nd February 2023
Part 1 of our interview with NearForm’s Kurt Milam and Lefteris Paraskevas
The digital health sector is growing rapidly. Statista notes that its revenue was $42.20 bn in 2017, with that figure rising to $145.76 bn in 2022. The sector is predicted to reach a revenue of $256.30 bn in 2027 — that’s an increase of 507% in ten years.
NearForm has partnered with a variety of organisations in the digital health sphere, from disruptive startups to global businesses. We’ve delivered a range of solutions for our clients, including designing, architecting and building scalable telemedicine platforms, as well as bringing ground-breaking products to market in record time.
We sat down with two of NearForm’s experts, Delivery Architect Kurt Milam and Technical Director Lefteris Paraskevas, to discuss how we use our expertise in digital health to help take businesses to the next level.
This is part one of our interview. It talks about the common goals, tech challenges and unique constraints of digital health companies.
Who are you both and what are your backgrounds?
“My name is Lefteris Paraskevas. I’m a Technical Director. I’ve been with NearForm for the last year and a half. I’m based in Athens, Greece, although I spent a big chunk of my life in London.
“I joined NearForm a year and a half ago. Before that, I worked for various companies including Salesforce, Amazon and Capgemini. I started my career as a developer and transitioned to management about seven years ago.”
“My name is Kurt Milam. I’m originally from Texas, but I live in Austria now. I have 25 years of experience in tech, working in various software development, consulting and management roles.
“I’ve worked for a number of companies over the years — various consultancies, Ogilvy, the advertising company, Dell and a tiny dot com startup back in Texas. I joined NearForm a little over five years ago. My current title is Delivery Architect (tech lead), and I’m happy to be here today!”
What are some of the common goals of digital health companies in your experience?
Providing top-quality services that are accessible
“Some of the stuff that we hear a lot from our clients around digital healthcare is that they’re looking to provide top-quality services and make healthcare accessible to people. They want to make it available at times and places that really work for their patients, so they don’t have to travel all the way to the clinic.
“The clients often have a physical clinic and utilise the digital aspect to do some appointments for things that might not be life-threatening.
“And the other thing that’s very common is they’re trying to take better control of their data and to understand it better.
“Ultimately, they want to provide tailored, personalised experiences to their patients. Potentially use AI models, or anything in this arena, to make the experience very specific to what the history of the patient is — based on the first-party data that they have.”
“The data side of things is very important. For instance, gathering biological markers, combining them with blood tests and running them through a machine learning model in order to forecast the likely path of progression of a specific disease
“There are also healthcare providers who would like to keep an eye on what types of diagnoses are spiking in real time, so they can learn about different demographics and what sort of ailments they are suffering from. Or they may want to combine population data to help identify common comorbidities.
“Another common goal is to minimise data errors caused by manual entry, in order to keep mistakes and incorrect data from making it into patient medical files.”
What are some of the tech challenges that digital health companies face in achieving their goals?
Modernisation and understanding the tech that’s available
“Healthcare companies have one very, very, very important duty, which is to provide healthcare and help people. Bringing technical improvements can make a huge difference to this and modernising what they currently have is definitely a challenge for companies in the digital health sector.
“A company might have very specific hardware that they use to run analysis, such as blood samples. They then find it very challenging to bring this into a tool that’s more user-friendly, more up to the current technology trends.
“Understanding what technologies are available to them is also crucial. That’s where NearForm can really help, because we can help companies connect different parts of their business.”
Moving forward while keeping within regulatory boundaries
“In my experience, companies working in the healthcare space tend to lag behind other industries when it comes to delivery. So, a big part of our role in working with digital health companies is to help them streamline their delivery process. How can they improve or implement continuous integration and deployment while complying with strict regulations and guaranteeing the highest level of quality? How should they test? How should they deploy services to the cloud? These types of things.
“Some of that stems from strict regulatory burdens that are a reality for most medical companies. These tend to slow companies in the industry down when it comes to adopting new tech and processes.
“So I think it’s an industry that tends to move rather carefully and slowly, in order to stay within the bounds of regulations. This leaves them in a spot where they often need help to get up to speed as quickly as possible, while working within the framework of regulation and quality concerns.”
Data security and data privacy is vital
“I would say that data privacy and security is more important than anything else and this underpins the challenges that digital health companies face.
“It’s not very easy for people to just build small tools to help these companies if they’re not really tested, if they’re not really assessed — we can’t just implement any other solution that will work for an e-commerce website because we’re talking about really, really sensitive data there.
“There might be some key players that have developed a very, very important tool that many people use, but it’s pretty old and this means it’s very hard to then transition to something more modern. You may build something modern, with a fancy UI, but you have to be certain it meets all the other requirements which are even more important than this.”
Focusing on quality is essential
“There’s obviously also the fundamental concern about maintaining patient well-being. That’s a crucial reason to really focus on quality and take things slowly.
“We don’t just need to make sure we tick off all the boxes in the regulatory compliance sheets. We must also do our best to ensure that our product doesn’t cause actual harm to any patient.”
What are the unique tech constraints that are faced by companies you’ve worked with?
Regulatory requirements aren’t universal
“There’s not one set of regulations across the globe. So any company that wants to expand its offering beyond a single country usually has to really understand the regulatory requirements of the region they want to go into.
“Certain regions are a bit more loose. Certain regions are a bit more strict. There are some certain golden standards that you can use from some countries, and you can always assume this is one way to do things. But then every country you need to adjust to.
“This means deploying to a cloud solution might not always be easy, or feasible, or aligned with certain requirements. For example, there might be data residency requirements. So if you want to build a solution for somewhere, let’s say AWS, or Google Cloud, or Azure, you need to build something more specific to meet the data residency requirement.
“You need to think a bit differently. For instance, accessibility is very important for every website, but it’s even more crucial here because we’re talking about offering healthcare to people.
“So the language, how you phrase things on the UI, is very important. To not mislead the patient, to make sure that you get the trust of the patient.”
Different regions can require different solutions
“There are other interesting concerns that don’t come up as often in other industries.
“For instance, there’s the idea of not wanting to have default values in forms. Because you don’t ever want to assume that the patient actually read that value and chose to submit it. Maybe they just skimmed over and finished the form as quickly as they could before submitting it.
“Where a value is required, you always start with an empty field, forcing the patient to actively choose an answer. The same goes for clinicians.
“There are also data residency regulations that we have to comply with in different countries.
“Some countries say, ‘We will not allow sensitive medical data about our citizens to be hosted outside of the country.’ That means we won’t be able to host a service in AWS, in the United States, then offer that service to patients or clinicians located in a country with a law like that. Because we’re not allowed to actually take that data out of the originating country and store it in the United States.
“That’s also pushed us into looking for solutions to this concern — how can we build a platform, for instance, that can be rolled out in multiple different countries around the world, that takes into account the fact that some of those countries will not allow us to bring medical data outside their border?
“And most cloud providers don’t offer their service in every country on the planet. Some countries will have AWS. Some will have Azure. Some will have Google Cloud or GCP. And some of them may not have any of those services.
“In a project I’ve been working on recently, we’re focusing on building a platform that the customer should be able to roll out into any country.
“We had to consider that as we were designing the architecture and choosing services — how can we build this cloud-based service without tying it down to a single cloud service, so it can be rolled out in any country, regardless of which cloud services are available there?”
Data access and logging is a major factor
“It’s very common for engineers that they need access to logs or metrics, just to be able to maintain their service and fix any problems.
“Even there, you need to be very careful. You cannot log any data from the patient. We need to potentially not log patient data, or obfuscate them, or find ways to not be able to allow developers to tie anything back to any specific person.
“So how do you build the reporting system, or dashboards for the business, that focus specifically on what they need, but don’t give them any more data than what is legal, or is even common sense?”
Balancing a product focus with doing the appropriate things for patients
“We’re very product-focused in NearForm and a lot of our clients try to take this approach as well. But digital health companies must walk a very fine line on product-focus versus being, potentially, a bit annoying for the end user. They also need to make sure they have the right data and the right processes.
“You might need to add an extra step in your registration or to gather more information from the patient.
“In other industries you might hear, ‘Oh, why do we need to do two clicks here, or three clicks here?’ But in healthcare, you might need to have these clicks, just to potentially have a disclaimer that says, ‘Before you answer, to make sure it’s not urgent, please call 911.’”