Hypo-METRICS: Hypoglycaemia - Measurement, ThResholds and ImpaCtS

King's College London602 enrolled

Overview

Hypoglycaemia or low blood glucose, and its fear are major barriers to achieving optimal glucose control. New technology, such as continuous glucose monitors (CGM), help to better identify hypoglycaemia and develop strategies to avoid it. These devices measure glucose in the skin, rather than in the blood, and provide information not only on how low glucose is, but also for how long. Recent studies showed that over half of episodes of low glucose with these systems are not recognised by people with diabetes, and even people without diabetes have sensor values that are below the current thresholds for hypoglycaemia \[ low blood glucose\] that we measure with traditional monitors. In this study, the investigators will evaluate the impact of symptomatic as well as asymptomatic episodes of low sensor glucose on a variety of clinical, patient-related and health economic outcomes such as mood, quality of sleep and productivity. The investigators will test different levels and durations of low sensor glucose to identify the one that best matches episodes that are symptomatic to best define hypoglycaemia using these systems. The investigators will also look at factors that influence this such as sleep or activity as well as diabetes management behaviours (such as insulin dosing, carb counting, etc). At the end of this study, the investigators will be able to provide a better definition of clinically relevant low sensor glucose readings that will help inform clinical as well as academic interpretation of CGM data.

Study Type

OBSERVATIONAL

Enrollment

602

Conditions

Diabetes Mellitus, Type 1 Diabetes Mellitus, Type 2 Hypoglycemia

Eligibility

Sex: ALLMin age: 18 YearsMax age: 85 Years

Medical Language ↔ Plain English

Inclusion Criteria: 1. Age 18 - 85 years 2. HbA1c 5 - 10% \[31 - 86 mmol/mol\] 3. Confirmed diagnosis of type 1 or type 2 diabetes 4. Using \> 1 injection of insulin / day or insulin pump. 5. Ability to provide written informed consent 6. Performing regular SMBG \[ \> 1 / day on a 4-week download\] . For those using flash or continuous glucose monitoring, this should be used at least 70% of the time. 7. At least 1 episode of hypoglycaemia \[ either biochemical or symptomatic\] in the last month 8. On stable therapy for at least 3 months. 9. Willing to complete study procedures including wearing the Fitbit and CGM devices and completing the EMA questionnaires on the uMotif app three times a day for 10 weeks ( we expect minimum 80% data completeness) Exclusion Criteria: 1. Concurrent conditions that can affect glucose readings \[renal impairment GFR \< 30 ml/min, hepatic impairment, untreated adrenal or thyroid insufficiency, as judged by the investigator. 2. Severe cognitive impairment or psychological illness that can impair performance of EMA tasks, visual impairment that will preclude use of the EMA or sensors. 3. Severe psychiatric / psychological illness including extreme fear of hypo- or hyper- glycaemia ( in the opinion of the investigator) 4. Pregnant or plans for pregnancy in the next 6 months 5. Use of automated insulin delivery systems such as closed loop or automated threshold suspend or predictive low glucose suspend insulin pumps. 6. Known allergies to adhesives required for the CGM systems 7. People who work regular night shifts 8. Any other condition which in the opinion of the study team would impair their ability to complete the study

Outcomes

Primary Outcomes

LIG optimum detection of hypoglycemia

To determine the Low Interstitial Glucose (LIG) parameters that have the optimum performance for detection and identification of patient-reported-hypoglycemia (PRH); 〖LIG〗\_PRH (h\_opt,t\_opt).

Time frame: week 10

Secondary Outcomes

Secondary outcome 1

To evaluate the impact of symptomatic and asymptomatic hypoglycaemia on different domains of Quality of Life (QoL) and health economic variables measured by a bespoke mobile phone application. We will use the definition of LIG identified in the primary objective, 〖LIG〗\_PRH (h\_opt,t\_opt), to evaluate the impact of symptomatic and asymptomatic hypoglycaemia on various outcome variables. To do this, we will divide the days into one of 4 categories LIG-PRH- :no hypoglycaemia LIG+PRH- :asymptomatic biochemical hypoglycaemia ( CGM false positive) LIG+PRH+ :symptomatic biochemical hypoglycaemia LIG-PRH+ :symptomatic without biochemical hypoglycaemia (CGM false negative) We will then compare the scores in different domains collected on the app ( eg reported mood, quality of sleep, alertness, productivity, time off work) between the 4 categories.

Time frame: week 10

Secondary outcome 2

To evaluate the impact of adding co-variates such as change of glucose for 30 minutes prior to the event and area under the curve to the threshold and duration parameters to see if this can improve the sensitivity and specificity of 〖LIG〗\_PRH (h\_opt,t\_opt),

Time frame: week 10

Secondary outcome 3

To evaluate the impact on these optimal parameters \[〖LIG〗\_PRH (h\_opt,t\_opt \] on Type of diabetes ( Type 1 with normal awareness, Type 2, Type 1 with impaired awareness) Awake vs sleeping ( as determined by Fitbit) Usual Mode of glucose monitoring (SMBG vs Flash vs CGM) Rate of 〖LIG〗\_PRH during the study

Time frame: week 10

Secondary outcome 4

To evaluate the impact of the rate of 〖LIG〗\_PRH over the whole study duration on Rate of Patient reported hypoglycaemia ( non-severe) Rate of severe hypoglycaemia Overall EQ-5D-5L scores WPAI ( productivity) PROMIS - (sleep quality) Hypo fear scores Baseline C-peptide levels

Time frame: week 10

Secondary outcome 5

To determine the parameters (threshold and duration) of low interstitial glucose (LIG) that best identify a pre-specified reduction/ change in QoL. \[ 0.07 reduction in EQ-5D-5L\] 〖LIG〗\_(EQ-5D) (h\_opt,t\_opt).

Time frame: week 10

Secondary outcome 6

To determine the parameters ( threshold and duration) of low interstitial glucose (LIG) that best identify a pre-specified health economic outcome. \[1.5 hour loss of effective work/ activity\] 〖LIG〗\_productivity (h\_opt,t\_opt).

Time frame: week 10

Secondary outcome 7

Determining the predictors of 〖LIG〗\_PRH by looking at variables such as mean glucose, variability, activity (step count and intensity).

Time frame: week 10

Secondary outcome 8

We will evaluate the impact of activity ( step count) on the risk of 〖LIG〗\_PRH the following night. A generalized linear mixed models (GLMM) with step count as the independent variable and the rate of LIG\_PRH as the response variable will address this research question

Time frame: week 10

Secondary outcome 9

We will try to understand some of the causes of hypoglycaemia. To do this we will look at the 4 hours prior to every available 〖LIG〗\_PRH (h\_opt,t\_opt) and PRH and describe the category of events (if also insulin data is available) by classifying them into those with Bolus within range \[ \< 10 mmol/l\] within 4 hours Bolus and correction \[ \> 10 mmol/l \] within 4 hours No bolus within 4 hours Activity \[ as identified by the Fitbit charge 3\] within 2 hours.

Time frame: week 10

Secondary outcome 10

Evaluating the effect of personality (assessed by DS-14) on number and severity of hypoglycaemia and the effect of hypoglycaemia on quality of life measures

Time frame: week 10

Hypo-METRICS: Hypoglycaemia - Measurement, ThResholds and ImpaCtS

Overview

Conditions

Interventions

Eligibility

Locations (1)

Outcomes

Primary Outcomes

Secondary Outcomes