FDA is looking for public comments on insulin bolus calculators. Here is the link to FDA’s full ask: Regulatory Science Considerations for Software Used in Diabetes Management; Public Workshop; Request for Comments
To comment go here.
If you need step by step instruction on how to comment, we have them here.
Make you comments unique by telling your personal story of life with diabetes. For tips on writing your personal story as an advocacy to see this link
1. How can patients and providers be confident that the insulin bolus values obtained from the calculators are accurate and appropriate for their use?
2. What information do patients and providers need about how a particular calculator works so that they may appropriately use the calculator for diabetes management?
3. How can FDA foster both innovation and safety of insulin dose calculators intended for use by healthcare practitioners?
4. How can FDA foster both innovation and safety of insulin dose calculators intended for use by patients?
Write you own responses borrow from ours below:
Patients and providers can be confident that the insulin bolus values obtained from the calculators are accurate and appropriate for their use, if the calculators are based on established insulin dosing algorithms and the blood glucose value driving the calculations is accurate. While not perfect, noting in diabetes care is, calculators do help minimize the mental math of diabetes and so improve patients likelihood of avoiding a dangerous insulin error.
FDA needs only minimal oversight of a calculator based on established algorithms used clinically. A slightly more robust consideration of innovative algorithms may be required, particularly as those become near real time manual artificial pancreas systems. FDA should expedite review of such innovation. Insulin is a very dangerous drug. Bolus calculators based on near real time information are highly valuable and promise significant improvements over more static approaches.
FDA should approve these systems and fully automated systems based on a clear understanding of the logic used for insulin action, sensitivity, active time, carb and other nutrient data, insulin on board and of course blood glucose. Patients and providers need to know the logic of how a particular calculator considers these items in a dosing calculation.
FDA can foster both innovation and safety of insulin dose calculators intended for use by healthcare practitioners in the clinical setting with a process that appreciates the importance of frequency of glucose measurement. In the acute setting, a prolonged period between finger sticks fails to appreciate the viability of glucose. The interaction of other medications and treatments can significantly have an impact on insulin response. Too often in the in hospital setting the glucose measurement and dosing are separated by enough time as to pose a risk. This delay is particularly important if single finger stick are used and the trend in blood glucose is not known. Sensing glucose continuously in the acute care setting would add the valuable information of trend to insulin blousing calculations.
FDA can foster both innovation and safety of insulin dose calculators intended for use by patients in the outcome setting by fostering innovation with speedy review of applications. FDA oversight is needed to help protect marginally engaged diabetes patients. These patients however stand the most to gain by through systems that help close the gap in care by more constructively dosing insulin.
One important path off innovation is bringing the expertise of the patient community into the process through open source solutions. This is a new avenue for device development. FDA can foster this innovation by continued public sessions and workshops for developers.
Dosing insulin to manage blood glucose is a significant amount of arithmetic. This mental math is in addition to diabetes considerations such as fat, fiber and carbohydrate composition of a meal, activity, stress and other factors that have an impact on glucose levels. Therefore even rudimentary calculator that are easy enough to use to be used, are better than rough estimating.
Calculators to be effective should be part of a collection of devices that share data. Interoperability between measurement of glucose, calculation of a bolus, insulin on board and the patient must be far more seamless than it is the current practice. Patents manually move data between meters, sensors, calculators and delivery devices. Often there is no tracking of time injection.
At their public workshop, FDA expressed concern about risks of the calculator is wrong. Since much of the process is established arithmetic the calculator itself is less likely to be a source of error than the data supplied for the calculation. A significant source of risk of calculator error is meters performance.
Meter accuracy is increasingly a concern. Evidence is mounting that the post-market performance of some devices is not delivering the standards of pre-market approval. The evidence suggests that the accuracy is deteriorating. Therefore, FDA should require specific labeling identifying meters that are sufficiently accurate both pre and post market to feed any bolus calculator.
I share the sentiment of Stephen Shaul’s comment at the FDA workshop that in a world bringing driverless cars to the road; computerization of insulin dosing should also be a reality.