The challenge of multiple daily insulin injections

Some improvements in diabetes care happen slowly, whereas others seem to happen very fast. For a number of reasons, in the last two years we have suddenly seen a dramatic change in our approach to insulin therapy. This has partly been due to the roll-out of continuous and flash glucose monitoring. This, combined with the enforced inability to see people face to face, meant we had to develop tools for remote monitoring and has resulted in a dramatic increase in data sharing between people with diabetes and clinicians, particularly with regard to glucose levels throughout the day. Instead of a meeting where a paper glucose diary with perhaps two to four readings per day is reviewed, we now have a wealth of detail about day-to-day glucose variation.

This change has shifted the conversation in clinic and, as we know, resulted in measurable improvements for the individual with diabetes. It does, however, highlight a gaping hole in the information that we need in order to give accurate advice: we know very little about the appropriateness and timing of insulin doses. At the moment, the information we have for an individual using insulin pump therapy will be very different to the information available for a person using multiple daily injections of insulin. Often we can only assume that insulin has, firstly, been taken and, secondly, been taken at an appropriate time and dose for that meal. Evidence would suggest that this assumption will be incorrect for a relatively high proportion of the time.

This paper by Susan Robinson and colleagues summarises our current knowledge about the timing of insulin injections for people with both type 1 and type 2 diabetes. The results are sobering but perhaps not surprising, with one or two doses of insulin being missed per person per week. The timing of bolus insulin in relation to meals is also important and, while we often assume in the clinic that the dose is being taken at the right point, the data would suggest this is probably not the case around 50% of the time.

These are the basics of insulin self-administration that we mostly assume are understood and practised correctly before considering factors such as matching the insulin dose to the carbohydrate content of the meal, the proportion of the previous insulin injection still active, correction doses to manage the current glucose level and the effects of previous or planned physical activity. When there is a problem with glucose control, we often assume it is these complexities that need addressing, without considering that just remembering to take the insulin correctly may be a challenge.

Technology is already helping, with automated systems to help calculate insulin doses and smart pens to remind about insulin injections, but in addition the consultation needs to evolve to deal with this new conversation. To best help the person with diabetes, we need to match the wealth of glucose data with equally accurate measures of insulin dose and timing. Hopefully this will come in the near future.

Click here to read the Digest.