Analyzing the output of the continuous monitoring survey – some thoughts on the data… #wearenotwaiting

Earlier this week I published the first set of results of the survey of CGM and Libre users that I undertook last year. There was a large amount of data from the survey, and the first report focuses on Type 1 outcomes. Notably, data referring to people with type 2 and the section on CCG attribution remains to be analysed.

The results backed up existing studies, showing that users of various systems typically saw a reduction in Hba1C, and overall this was an average reduction of 1.1% in the mean Hba1C level. Participants time in range went from a majority not knowing to a majority exceeding 50%. Those who didn’t use their systems full time saw less of an Hba1C reduction than those who did (t-tested using the Abbott data, where it’s better correlated whether someone is using a system full time based on finite lifespan of the sensor and expenditure on consumables).

Interestingly, fewer Abbott Freestyle Libre users reported a reduction in frequency and severity of hypoglycaemic events than CGM users, which is perhaps unsurprising given the lack of alarming on the Libre. For most systems, frequency reduction was reported by more than 60% of participants, whereas for Libre it was 53%, and in terms of severity reduction, for other systems more than 60% reported a benefit but only 39% of Libre responders did so. Whilst this is an obvious benefit and 39% seeing it is good, as a comparison with other systems, it’s not as good, even with cost neutrality compared to SMBG. Other than the lack of alarming already mentioned, I wonder whether this is also linked to the anecdotally reported variation in the Libre sensor readings.

Looking at how people were funded, comparing NHS users with paying users showed there was correlation between whether someone was funded and reduction in Hba1C, however, it needs to be borne in mind that most NHS funded users receive CGM as a result of impaired hypo awareness, so they may not be attempting to improve their management in such a way that presents with a lower Hba1C. Having said that, other research has shown that lower Hba1Cs tend to be correlated with fewer severe hypos, so they may still have benefited.

What was really notable was that those who responded to the survey generally had a better distribution of Hba1C than the UK NDA showed for type 1s, pre-use of a system, and a much better distribution post use of one. I’ve suggested that demographics of users would be a beneficial addition to the dataset, and I believe that educational achievement would need to be a part of that. I postulate that one of the main reasons that the Hba1C distribution in the participants pre-use of CGM is better than that of the general UK population is down to education and awareness of diabetes. If you are someone that has bought into a relatively expensive system such as Libre or Dexcom, it’s likely that you have already established that you need it through traditional management techniques and are looking to improve your results. (It’s also possible that you were hypo unaware and swinging from high to low in an uncontrollable manner that you are trying to address).

On that note about these being relatively expensive systems, the other point that I took away from the survey results was the amount of money being spent on consumables for these systems. The average sum, using a weighted average based on the distribution of the responses is £1,465. This is strongly driven by the number of Libre users responding to the survey, for whom full time use costs roughly £1,300 and, I suspect, the extension of other brand sensors by users to avoid having to replace them as frequently as the manufacturer suggests. Cost was the one issue raised most frequently by both those using system and those that have stopped.

Reading the written feedback that a fair number of the participants provided, it’s clear that the ability to see what’s going on and spot patterns is of enormous use to people when using these systems. This ties in with the feedback that other qualitative studies have given, and is intrinsically linked to the metabolic quantitative results that participants reported.

So what did I take from this? Those who responded were very much in favour of continuous data, but that was expected and the population that the survey set out to address. The average amount of money being spent was also pretty revealing. For anyone on the UK average salary, the average cost of the consumables works out at around 5% of their income, which is likely to be way too high for most. As a result, the user base of this technology in the UK (and probably in other countries as well) is likely to be wealthier and more highly educated, simply because they’re the only ones who can afford it, and this harks back to the discussions about access that ATTD raised.

It demonstrates clearly the reasons why people are going out and trying to extend the life of glucose sensors (to 6-8x recommended life in some cases) and replacing the batteries in transmitters. People value the improved quality of life and better metabolic indicators, but can’t really afford to pay for it constantly. And therein lies the rub.

Ultimately, to get the metabolic benefits that can be achieved, firstly you need to want the tools to get that and secondly you need to be able to afford them. And there it is. Whilst we do, at least, get insulin and delivery systems without having to pay anything additional, if you want the more advanced tools that provide better outcomes? Well you’ll just have to work out how to pay for them, at least for the time being.

1 Comment on Analyzing the output of the continuous monitoring survey – some thoughts on the data… #wearenotwaiting

Leave a Reply

Your email address will not be published.