A new metric?

As we continue our prep work for the upcoming ADA conference which begins the end of next week it dawned on us that as much as we don’t want to overcomplicate things, we need a new metric. At ADA one metric which is sure to garner attention is TIR or Time in Range, TIR combined with HbA1c is becoming the new standard for defining what is and what is not good control. Thanks to growing usage of CGM we are learning more and more that HbA1c long the gold standard for defining control by itself is an incomplete measure.

Now before we go any further it should be noted that HbA1c is just fine for patients who don’t use insulin. Yes it would be great if these patients adopted CGM and used it daily but that is pure fantasy no matter how cheap CGM becomes. In the real-world patients want simple and HbA1c is about as simple as it gets anything below 7 is good anything above 7 not so good.

When it comes to CGM and non-insulin using patients it’s all about three things;

1. Discovery – let’s see what the heck is going on here.

2. Follow Through – are we on the right path with any changes we made.

3. Maintenance – is everything still working, or do we need to make additional changes.

CGM will be a valuable tool for these patients but it will not be used on a daily basis.

Yet in the insulin using population CGM is quickly becoming the standard for measuring glucose whether the patients uses a pump, pen or syringe. Thanks to insulin dosing algorithms combined with accurate, reliable and affordable CGM’s it will be easier than ever for these patients to dose their insulin. It will also make TIR another metric that is measured along with HbA1c.

In the future we can see a world where a patient and their healthcare team make a decision on which insulin delivery system they use based on these metrics. Of course cost will always play a role however in our world with lots of me-too products metrics such as TIR could be the determining factor in selecting say one Tyler over a competing Tyler.

To us this is just one more step towards the consumerization of healthcare providing the patient with as much information as possible so they can make an informed decision. To a certain extent this is happening already on Facebook and the many blogs devoted to diabetes. Patients share their experiences which helps other patients make a more informed choice.

However as helpful as these users’ experiences can be one key metric is missing from the decision-making process, reliability. Keep in mind that hybrid closed loop insulin delivery systems and Tyler are a collection of devices which talk to each other. Should any link in this chain fail or malfunction the entire system falls apart. While we aren’t big fans of the 670G from Medtronic when the damn thing works as designed it does a good job. The problem with the 670G is reliability in particular the reliability of the sensor.

It would be very helpful if patients knew how often the parts of the system fail or malfunction. Do they work 90% of the time or just 50%? To some extent this information can be found on the FDA’s MAUDE database but let’s be honest here how many patients even know about the MAUDE database let alone how to use it. The MAUDE database isn’t like going to Amazon and then reading reviews or seeing what the average star rating is.

This is what’s needed today for insulin delivery systems some sort of metric which tells the patient just how reliable the system functions in a real-world setting. Listen we aren’t delusional and understand the complexity of making this happen. But as Momma Kliff used to say just because something is difficult to accomplish doesn’t mean it should not be done. Keep in mind that these systems deliver not just a life sustaining drug but also a lethal drug when improperly dosed.

It seems to make sense that a patient would prefer a system that worked 90% of the time over one that only works 50% of the time. This will become even more important as we move closer and closer to mix and match systems. As we have stated many times the key components of these systems are not the hardware which delivers the insulin. Hardware whether it’s a pump or pen is a commodity. The key components are the CGM sensor and insulin dosing algorithm.

By our way of thinking the CGM sensor is the most critical as the insulin dosing algorithm makes its calculations based on this data. Put garbage in and you’ll get garbage out. We hate to be redundant but the CGM sensor is the straw which stirs the insulin delivery drink. If that sensor does not work, if it does not deliver an accurate reading IF IT’S NOT RELIABLE it does not matter which pump, pen or algorithm is attached to the system.

The fact is all these systems even the 670G are great WHEN they work as designed. The problem comes in when they don’t work as designed. It’s about time everyone paid closer attention to this fact.