Metrics that matter for a digital diabetes management program
11/13/2018 by mySugr
Mobile apps are a fast paced technology that require ongoing iterations. At mySugr, we’ve learnt this first hand. A small change in something as “banal” as our welcome email can reverberate in unexpected ways throughout our 1.5 million user base.
For better or worse, this means technology like ours doesn’t translate well into randomized clinical trials, the gold standard of prospective trials. By the time the publication date rolls around, we’re likely to be an entirely different animal.
The upside is that we have direct access to real world evidence. With our collaborator ProSciento, a leading metabolic clinical research organization, we can track and analyze the effects and changes we see among our more than 1.5 million users. Instead of pulling subjects into a structured experiment, we get an honest, organic look at how people from different walks of life manage their diabetes day-by-day.
So, what have we learned? At the 18th Annual Diabetes Technology Meeting (DTM), which took place in Bethesda, Maryland, ProSciento and mySugr presented two posters:
A diabetes care program drives real world results
Together with ProSciento, we looked into the impact of the mySugr Bundle among our users. The mySugr Bundle is a subscription kit containing unlimited test strips, automatic supply refills, personalized coaching, and more. It is available to patients directly (out of pocket) and through health insurers and provider systems in Germany and the United States. Learn more about our digital population health management solution here.
We looked into how people interact with the app just after registering, in the two months prior to receiving the mySugr Bundle and the two months after the bundle has arrived. We included data from people who measured their blood glucose levels on average three or more times per day during each period. Given the strict time frames and test frequency requirements for inclusion, and the recent launch of the mySugr Bundle in the U.S. market, the result was a pool of 52 participants with type 1, type 2, or latent autoimmune diabetes (LADA).
The results are very encouraging. The population was already doing well and their diabetes management improved further:
On average, participants’ blood glucose levels dropped from 154±55 mg/dl to 138±50 mg/dl
Translated into HbA1c (by calculating the “estimated A1c”), this shows a mean drop from 6.7% to 6.3%
This drop of 0.4% in an already “well controlled population” exceeds the 0.3% threshold for a therapy, device or medication to be seen as relevant
There was also a significant increase of 20% in testing frequency among bundle users.
The most startling results came through an analysis of participants who were “above the median” (n=26), with a mean blood glucose (BG) greater than 154 mg/dl (or > 6.7% eA1c). These users experienced a dramatic improvement in BG control (1.1%), from 186±72 mg/dl (7.7% eA1c), to 147±57 mg/dl (6.6% eA1c).
As lovable as Netflix
The mySugr Bundle was initially designed for people with type 1 diabetes, but over time we have seen strong uptake by the type 2 population. Members of this latter group who use the mySugr Bundle were analysed in a separate study, also presented at the 2018 Diabetes Technology Meeting, based on the results of a Net Promoter Score survey.
The key question we asked through the Net Promoter Score (NPS) survey was; “How likely are you to recommend mySugr Bundle to a friend?” NPS is a tool that, using social mechanics, puts the question of love in context. Even if a user is “satisfied,” that doesn’t mean he or she would recommended it to a close friend. It’s a commonly used tool that can lead to deeper questions about why people love the product, as well as what needs to be worked on.
As the score was calculated, we were surprised to see an NPS of nearly 70; on par with the likes of Netflix and Apple. Around 86% of respondents also reported being happy with the service they received through the remote certified diabetes educator (CDE) coaching.
[caption id="attachment_12577" width="684"]NPS graphic comparing mySugr with Netflix, Inc. and Apple.[/caption]
[caption id="attachment_12587" width="1080"]Graphic showing high satisfaction rate with coaching.[/caption]
We believe this topic is critical to the success of any digital diabetes management program. The success of mobile health tools often comes down to psychology – the device and the app are an extension of the user, which means they have to perfectly sync with the lives of people with diabetes.
The second question in this field is on access, and as you can see that is a field where we are taking dedicated steps and giant leaps. Click here to learn about how we are bringing digital diabetes management to rural communities.
Virtual diabetes care can produce real world outcomes
There was a lot to celebrate in the two posters, including the fact that mySugr’s digital health solution can make a clinically relevant impact in diabetes management. By looking at real world evidence, we showed tangible results from our virtual diabetes care. All of this was achieved by engaging a patient community that can spread the gospel of what brought them there.
We’ll keep you posted on the future learnings in this field, and look forward to writing our next update!
Make diabetes suck less! It's our mission, our motto, our way of life. We're thrilled you stopped by. Learn more about our products. Learn more about our company.