A method for estimating the risk of a future hypoglycemic event for a subject with a standing insulin regimen, wherein the standing insulin regimen comprises one or more types of insulin medicament dosage regimen, wherein each of the one or more types of insulin medicament dosage regimen comprises a type of insulin medicament defining one or more types of insulin medicaments, the method comprises:
- using a device;
- the device comprises one or more processors and a memory, the memory storing:
- one or more types of insulin medicament, and for each of the type of insulin medicament a duration of action profile for predicting the insulin remaining in the subject as a function of time and characterized by a duration of the respective insulin medicament, wherein the type of insulin medicament can contribute to insulin on board if a corresponding injection event is having a time stamp within the duration,
- a historical data set comprising historical insulin on board data for the subject as a function of time within a historical time course, historical glucose measurements of the subject and corresponding timestamps obtained within the historical time course, wherein the historical glucose measurements and the corresponding timestamps, are structured to allow a derivation of a first derivative with respect to time and thereby to obtain an evaluated historical glucose concentration, and an evaluated historical rate of change of glucose as a function of time, and wherein the historical time course comprises a plurality of historical time periods in adherence, wherein the subject has been in adherence with the standing insulin regimen,
- a plurality of historical time period records obtained from the historical data set wherein each of the historical time period records comprises an identified historical metabolic state of the subject, a historical blood glucose event, wherein the historical metabolic state can be identified by the historical blood glucose event, wherein the historical blood glucose event can be identified as a pattern in the historical glucose measurements of the historical data set, a respective type of metabolic state, and a corresponding historical time period in adherence with the standing insulin regimen, wherein the corresponding historical time period in adherence is one of the plurality of historical time periods in adherence;
- the memory further storing instructions that, when executed by the one or more processors, perform a method of:
- obtaining a first data set from one or more insulin injection devices used by the subject to apply the standing insulin regimen, the first data set comprising a plurality of insulin medicament records over a time course following the historical time course, each respective insulin medicament record in the plurality of medicament records comprising:
- (iv) a respective insulin medicament injection event including an amount of insulin medicament injected into the subject using a respective insulin injection device in the one or more insulin injection devices, and
- (v) a corresponding electronic injection event timestamp within the time course that is automatically generated by the respective insulin injection device upon occurrence of the respective insulin medicament injection event,
- (vi) a respective type of insulin medicament injected into the subject from one or more types of insulin medicaments;
- obtaining a second data set, the second data set comprising a plurality of autonomous glucose measurements of the subject and, for each respective autonomous glucose measurement in the plurality of autonomous glucose measurements, a glucose measurement timestamp representing when the respective measurement was made, wherein the glucose measurements and the corresponding timestamps, are structured to allow a derivation of a first derivative with respect to time and thereby to obtain an evaluated glucose concentration and a rate of change of glucose as a function of time;
- using the second data set to obtain a current blood glucose event associated with the subject, wherein the current blood glucose event can be identified as a pattern in the autonomous glucose measurements of the first data set, wherein the current blood glucose event identifies the beginning of an evaluation period relating to an evaluation of a current metabolic state of the subject, wherein the current metabolic state is having a respective type of metabolic state;
- responsive to an identification of the current blood glucose event associated with the subject at a given time, evaluating the hypoglycemic risk at an evaluation time within the evaluation period by:
- using the second data set to obtain the type of metabolic state of the current metabolic state;
- using the first data set to obtain an evaluated insulin on board of the subject, wherein the evaluated insulin on board is calculated from a total amount of insulin medicament injected into the subject indicated by the medicament records in the first data set having injection event timestamps that are within the duration of the respective insulin medicament to the evaluation time, and therefore contribute to the insulin on board, wherein the evaluation utilizes the duration of action profile of the medicament records indicating the total amount of insulin medicament injected,
- using the second data set to obtain:
- (i) an evaluated glucose concentration at the evaluation time,
- (ii) an evaluated rate of change of glucose at the evaluation time;
- responsive to the evaluated rate of change being negative:
- using the plurality of historical time period records to obtain:
- (iv) a reference historical metabolic state of the subject, by selecting a historical time period record, comprising a historical metabolic state having the same type of metabolic state as the current metabolic state, wherein the selected historical time period record defines a reference historical time period record,
- (v) a reference historical time period corresponding to the reference historical time period record,
- (vi) a reference historical time, by selecting a time within the reference historical time period, wherein a progression of the reference historical metabolic state at the reference historical time is comparable to the progression of the current metabolic state at the evaluation time;
- estimating a hypoglycemic risk measure wherein the hypoglycemic risk measure is an increasing function with the number of binary risk expressions being true, wherein the binary risk expressions are:
- (iv) the evaluated glucose concentration is smaller than an evaluated historical glucose concentration evaluated at the reference historical time,
- (v) an evaluated rate of change of glucose is numerically larger than an evaluated historical rate of change of glucose evaluated at the reference historical time,
- an evaluated insulin on board evaluated at the evaluation time is larger than an evaluated historical insulin on board evaluated at the reference historical time.