Platforms and Services

To me, it's quite clear that a complete system focusing just on diabetes data management won't succeed in the modern world. Rather, any system should be a part of a greater ecosystem of all health services. It is equally clear that a generic healthcare data management system cannot meet all the needs of different patients. This blog post elaborates on these thoughts.

Do one thing, and do it well

The first thing to take into account is focus. A system built to enhance diabetes treatment should really concentrate just on diabetes treatment. No generic healthcare solution will ever be as effective and useful as a system specifically tailored for diabetes treatment and designed with the needs of people with diabetes and their healthcare professionals in mind.

Being able to concentrate on diabetes treatment means that someone else should take care of things that are generic to all of the services, such as data storage and sharing, sign-on mechanisms, authentications, and so on. And that someone else should concentrate on that task, implementing such a generic platform as well as possible, for all those services to use.

You're not alone

There will be several services for managing diabetes data. No service will be perfect for all people with diabetes. People have different challenges with their condition, and they should be able to select the tools that best fit their individual needs. There also needs to be competition between services. That drives development and pushes each of the services to become the best they can be. It should also be possible to move from one solution to another, as your life changes. 

We should also remember that it's never just about diabetes. All people with diabetes are individuals, and the properties that describe their diabetes are never adequate to portray a holistic picture of their life. They have other illnesses and conditions, and that information needs to be taken into account when making adjustments to their treatment. I really don't think offering a complete end-to-end system for just diabetes data management works.

Besides, why would one want to implement all that generic functionality like data storage all over for each service? And even more importantly, why would a doctor need to sign on to several applications when meeting a customer? Healthcare organizations are very well aware of these issues nowadays.

There really is a genuine need for those platforms that store and manage all that data for all those services. And all the services really need to be integrated and talk to each other.

You're either a platform or a service

In my view, one should really make a choice. Either you're implementing a platform with open interfaces really well, or you're providing well designed services on top of such a platform. It's really difficult to be making both, while still keeping the interfaces open for all possible services, in a way that's fair and accessible for everyone.

There are many companies out there working really hard to make life easier for people with diabetes. I hope none of them think they're able to be the only solution making that possible, and really encourage them all to evaluate first whether they're offering a platform or a service. In the case of a platform, they should ask how they can best allow all third party services to operate on them, and not just limiting to diabetes services. A big question is also whether they are really simplifying the current mix of existing platforms or just adding to the complexity. In the case of offering a service, the companies should see how they can really add value to the whole ecosystem.

Would I share my personal health data?

My previous post included some of my personal health data. The Word document produced by the Dexcom software even includes the word Confidential in the footer. As I'm creating a new health related product, I must pay special attention to data privacy. I also need to think about different feelings people may have towards sharing their data.

Some time back I learned that Mendor, a Finnish startup company focusing on diabetes treatment, were installing their software product, the Mendor Balance, to the diabetes clinic I visit. Unfortunately that software did not support my glucose meter. However, when I contacted Mendor about the matter they said they'd be happy to build the support if they just got some example data from an actual device. So I sent them the logs from my meter, including real measurement data over several months. I had no concern over what they might think of me as a diabetic or as a person, based on that data. They just wanted to get some sample data to implement a piece of software.

Whether I feel comfortable sharing my personal health information depends on how I believe that information is being used.

Of course, in direct contact with a person, sharing information is much easier. I've had almost no concern explaining everything about my life to the doctors and nurses who have helped me with my diabetes. The more information they have, and the more accurate that information is, the better they can guide me.

I think I've shared even more with the people in my peer support group. There it's all about finding the fun in living with diabetes, we all exaggerate our foolish behaviors and the mistakes we've made. Absolutely no fear of judgement or patronizing. I've shared more than just data, I've shared stories, big and small.

That's one thing in itself, isn't it? I feel much easier disclosing sensitive information when I can annotate it and share some additional background to it. I'd be reluctant to share just some cold piece of data, just describing a tiny tiny piece of myself, as I wouldn't like people (or computer algorithms for that matter) forming their opinion of me just based on that data.

I must admit, for the previous post, out of seven days of monitoring data I selected the day with the nicest graph. The graph for the whole week is much less stable. It's just that I don't know all of you and you don't know me well enough, for me to feel comfortable sharing the bad days. I'll probably get more comfortable with that as well, once I get enough background context out in this blog.

Whether I feel comfortable sharing my personal health information depends on

1. how I believe that information is being used
2. whether I believe that piece of information is both adequate and meaningful for that purpose.

Of course, there is also the whole issue of trust in general. I'll probably contemplate that in another post.