The MiaoMiao study: can do-it-yourself continuous glucose monitoring technology improve fear of hypoglycaemia in parents of children affected by type 1 diabetes?

Abstract

Purpose

Type 1 diabetes (T1D) is one of the most common chronic diseases of childhood and comes with considerable management and psychological burden for children and their families. Fear of hypoglycaemia (FOH), particularly nocturnal hypoglycaemia, is a common worry. Continuous glucose monitoring (CGM) is a tool that may help reduce FOH, as well as reduce overall diabetes burden. However, CGM systems are expensive and often not publicly funded or subsidised. MiaoMiao (MM) is a novel relatively affordable third-party add-on technology to intermittently scanned CGM (isCGM). MM allows users to convert their isCGM to a form of “Do-it-yourself” (DIY)-CGM. Our hypothesis is that MM-CGM will result in significant reduction in parental fear from hypoglycaemia. The primary objective is to determine the impact of real-time DIY-CGM on parental fear of hypoglycaemia using Hypoglycaemia Fear Survey (HFS).

Methods

This is a multisite randomised cross-over study of 55 New Zealand children (ages 2–13 years) with established T1D and current users of isCGM (Abbott FreeStyle Libre). DIY-CGM will be compared to usual care with isCGM. Participants will be randomised to either arm of the study for 6 weeks followed by a 4-week wash-out period before crossing over to the other study arm for a further 6 weeks.

Discussion

The results of this study will provide much needed clinical trial data regarding DIY-CGM effectiveness in reducing parental FOH, as measured by HFS, as well as various other secondary outcomes including traditional glycaemic metrics, and child and caregiver sleep. The trial was registered with the Australian New Zealand Clinical Trials Registry (ACTRN 12619001551189) on 18 November 2019, and the World Health Organisation International Clinical Trial Registry Platform (Universal Trial Number U1111–1236-9189).

Background

Type 1 diabetes (T1D) is one of the most common chronic diseases in childhood [1]; resulting in approximately 100,000 new cases annually worldwide among children <15 years [2]. Intensive diabetes management is at the core of modern T1D care [3]. This intensive management, aimed at achieving near normal glycaemic control, has been shown to reduce the risk of long-term complications [4, 5], but comes at a price. A three-fold increase in severe hypoglycaemia was seen in the Diabetes Control and Complications Trial (DCCT) in those who received intensive management compared to conventional therapy, mostly occurring during sleep [6]. Moreover, severe hypoglycaemia remains a big challenge for people with T1D across their life span [7]. Given the adverse effects of hypoglycaemia that range from anxiety and irritability, to seizures and even death, it is therefore no surprise that fear of hypoglycaemia (FOH), particularly at night, is a common problem among children with T1D and their parents [8–12]. This fear may impact efforts to obtain healthy glycaemic control [8–12], conduct exercise [13], and lead to less healthy dietary habits such as increased carbohydrate intake or snacking at night [14, 15].

Glucose monitoring, a vital part of intensive diabetes management, is a crucial tool in appropriately managing hypoglycaemia [3]. Increased glucose monitoring is also associated with improved long-term glycaemic control; as measured by HbA1c. [16, 17]. However, in childhood, parents play a major role in monitoring (especially overnight) [18] which consequently adds to the overall management burden and can have negative effects on quality of life and sleep quality for both parents and their children [19].

Continuous glucose monitoring

Newer glucose monitoring technologies, in particular continuous glucose monitoring (CGM), measure interstitial glucose as opposed to traditional finger-prick blood glucose measurement. CGM offers various potential advantages over capillary glucose monitoring including additional information on glucose trends as well as glucose threshold alarms which can be set and adjusted for (both hypo- and hyperglycaemia), thus potentially reducing the need for glucose testing overnight [20]. CGM has been shown to improve glycaemic control and reduce hypoglycaemia rates [21–23]. However, CGM is expensive and there are barriers for universal access.

Intermittently scanned CGM (isCGM) is a cheaper technology, which also detects interstitial glucose and provides many of the “on-demand” benefits of CGM. While isCGM requires no finger pricks to monitor glucose levels, it does not provide consistent recording of glucose trends. Subsequently, isCGM does not offer the glucose threshold alarms available when using CGM. Recently, the FreeStyle Libre-2 system which offers limited alarms was released; however, it is not yet available in most countries. In this regard, a recent study has suggested that CGM is superior to isCGM for improving time in target glucose range and preventing hypoglycaemia [24]. The prevention of hypoglycaemia is a significant advantage of CGM, and has the potential to be lifesaving. However, these benefits come at a substantial increase in cost – with isCGM costing approximately $US1600/year vs $US3200–6400/year for CGM [25]. Finding ways to provide greater and cheaper access to this expensive technology is vital if the majority of the population are ever to access the benefits.

Do-it-yourself CGM

#WeAreNotWaiting is a patient-led diabetes technology innovation and advocacy movement that was developed in response to the slow rate of development of patient-centred digital diabetes technologies [26–28]. As DIY projects, users are responsible for setting up their own systems with limited online support provided by the wider DIY/open source diabetes community. In contrast to isCGM, which is factory calibrated and cannot receive finger prick calibrations, these DIY systems utilise finger prick calibration, which could be considered a positive feature for accuracy and overall performance, although confirmatory data is currently lacking.

Recently, a third-party device (MiaoMiao (MM)) (MiaoMiao version 2, Smart Reader, Shanghai High Brilliant Health Technology Co. Ltd., China), has been introduced into the market. The device can be purchased online for $US169 as a one-off cost, in addition to ongoing isCGM costs [29]. This technology (MM-CGM) uses a Bluetooth transmitter that can be combined with isCGM to offer many of the advantages of currently available commercial CGM yet at a significantly reduced cost. MM is designed to be placed over the standard isCGM sensor; the device then uses Near-Field Communication (NFC) to read raw data from the isCGM sensor. Using Bluetooth™, MM transmits data to a paired smart device (usually the user’s phone) bypassing the official Abbott™’s algorithm (Fig. 1) [26]. A number of CGM applications are available to process the raw data from isCGM; these apps include proprietary third party CGM apps, Glimp [24] or Tomato [26], or open-source CGM applications developed by the wider diabetes community, Spike or xDrip+.

Fig. 1.

Illustration of MiaoMiao Continuous Glucose Monitoring setup

Driven by lower cost, patients and their families are adopting MM-CGM as an affordable CGM alternative; however, evidence of effectiveness and safety is limited. Recently, a qualitative study on the patient/family experience of using a MM-CGM has been published, suggesting an overall positive experience [26]. With more families adopting MM-CGM, research is needed to fill the gap in evidence regarding the impact of using DIY-CGM on various aspects of T1D management, including psychological and glycaemic outcomes. If research shows that this more affordable add on technology can improve patients’ experience and reduce fear from hypoglycaemia DIY-CGM may be able to take on a greater role in management options for people impacted by diabetes.

Aim, hypotheses and objectives

This study aims to investigate the effectiveness of an isCGM system converted to CGM (MM-CGM) vs standard isCGM to reduce parental FOH.

Secondary objectives include investigating the impact of the MM-CGM on glycaemic variables and various psychological outcomes such as child FOH, quality of life, diabetes treatment satisfaction, and parental and child sleep quantity and quality (Table 1).

Table 1.

Primary and secondary outcomes

Primary outcome

• Parental fear of hypoglycaemia (measured by Hypoglycaemia Fear Survey (HFS).

Secondary outcomes

• Glycaemic variables (each study arm 6 weeks):   a. Time in target range (3.9–10 mmol/L).   b. Time spent >10 mmol/L.   c. Time spent >13.9 mmol/L.   d. Time spent in hypoglycaemia (glucose <3.9 mmol/L).   e. Time spent in hypoglycaemia (glucose <3.0 mmol/L).  • Psychosocial variables:   a. Child fear of hypoglycaemia which will be measured by Hypoglycaemia Fear Survey (HFS) for children (age 6–13) [34, 54]   b. Quality of life (QOL) collected via the Pediatric Quality of Life Inventory (PedsQL™) and the PedsQL™ Diabetes Module. The later will assess children’s diabetes-specific health-related quality of life over the past month. [42, 55]   c. Quality of life (QOL) for parents will be collected via PedsQL™ Family Impact Module version 2.   d. Diabetes Treatment Satisfaction Questionnaire, (DTSQ) which measures the degree of satisfaction before and after using MM-CGM.   e. Parent and child sleep timing, quantity and quality will be measured objectively by 7-day actigraphs and sleep disturbance and impairment by using the, Patient-Reported Outcomes Measurement Information System (PROMIS) questionnaires.   f. MiaoMiao Continuous Glucose Monitoring user experience.

MM-CGM MiaoMiao Continuous Glucose Monitoring

The overall hypothesis is that, in comparison to isCGM, MM-CGM will result in a clinically significant reduction in parental fear from hypoglycaemia of at least 8 points on the hypoglycaemia fear survey [HFS] scores [30]. Additionally, it is hypothesised that MM-CGM will result in other positive impacts on secondary outcome measures such as quality of life, sleep, diabetes management and T1D treatment satisfaction, compared to isCGM.

Methods

Study design

This project is a multisite randomised 17-week cross-over study. This comprises a 1-week run-in period for baseline observations prior to study commencement. Families will be randomised to the order in which they receive the two interventions: MM-CGM or standard isCGM. As shown in Fig. 2, participants will receive their first intervention for six weeks, followed by a wash-out period of four weeks, to reduce carry-over effects of MM-CGM. Participants will then receive their second intervention for an additional six weeks.

Fig. 2.

MiaoMiao study design. CGMS, Continuous Glucose Monitoring System; isCGM, intermittently scanned Continuous Glucose Monitoring.

Recruitment

During routine clinical visits, eligible children already using isCGM will be identified by their usual paediatric endocrinologist /diabetes specialist and invited to participate. Participants will be recruited through diabetes services in district health boards (DHBs) across New Zealand (Southern, Capital and Coast and Auckland DHBs) encompassing an estimated total population of 2.4 million people [31].

The recruitment process flowchart is illustrated in Fig. 3. Participants must meet all of the inclusion criteria to participate in this study (Table 2).

Fig. 3.

Flowchart of the MiaoMiao study. CGMS, Continuous Glucose Monitoring System; isCGM, intermittently scanned Continuous Glucose Monitoring.

Table 2.

Inclusion and exclusion criteria for participation

Inclusion criteria

1. Aged 13 years and under with parent/caregiver involved in diabetes management.

2. Already using isCGM technology with no restrictions based on insulin regimen.

3. Diagnosed with T1D for at least 6 months with no restrictions based on insulin regimen, using either insulin pump or multiple daily injections.

4. No restrictions placed based on glycaemic control.

5. Plans to continue with routine clinical care during the whole period of the study.

6. Intention for continuous use of isCGM during the whole study period (17-week).

7. Currently residing in and expecting to remain in regions served by the District Health Boards for whole period of the study.

8. Ability to understand study procedures, including English language proficiency, and to comply with them for the entire length of the study.

Exclusion criteria

1. Already using MM-CGM or another CGM product (other than isCGM).

2. Any severe diabetes related complications (nephropathy on treatment, retinopathy with associated visual loss – milder degrees will not be excluded).

3. Other severe uncontrolled medical or psychiatric co-morbidity/severe mental illness.

4. Participation in another device or drug study that could affect glucose measurements during the study period.

5. Inability of legal guardian to give written informed consent and/or provide day-to-day diabetes management input.

6. Plan to leave study regions prior to study completion.

isCGM intermittently scanned Continuous Glucose Monitoring, MM-CGM MiaoMiao Continuous Glucose Monitoring, CGM Continuous Glucose Monitoring

Study procedures

Screening and enrolment

Clinicians will provide potentially eligible patients and their parents/guardians with an overview of the study and explain the basic inclusion and exclusion criteria. If eligibility is confirmed, the Participant Information Sheets and Consent/Assent Forms will be given to potential participants. Written consent will be obtained by another member of the research team during the first study visit (visit 0). The visits will be completed through face-to-face meetings or online via Zoom videoconferencing platform (Zoom Video Communications, Inc., San Jose, California, USA). The parent/caregiver (the primary diabetes caregiver) will be provided with their own Participant Information Sheet for the parent/caregiver-specific study, if their child is eligible. All demographic data will be collected during the baseline visit (visit 0). Also, during visit 0 participants will be asked to wear the actigraphs sleep device for one week.

Participant will be scheduled for Visit 1 to answer the questionnaires and for the randomisation, at the end of the one-week run-in period.

Randomisation

After a one-week run in period of standard isCGM for all participants, eligible consenting children/families will be randomly allocated at visit 1 to the order in which they will receive the treatments- either intervention (MM-CGM) or control (using isCGM only) only after completing all baseline questionnaires. Randomisation (stratified by study site) will be conducted using Sealed Envelope; http://www.sealedenvelope.com/.

Study groups

The standard isCGM (Abbott Diabetes Care Ltd., Witney, UK) which will be in use by all participants at study commencement consists of an easy to apply, wear and use interstitial glucose sensor and glucose reader. The sensor is worn in the upper arm for up to 14 days and requires no blood tests for calibration. The FreeStyle Libre reader held over the sensor to obtain an instant reliable, and accurate glucose reading [32]. The stored glucose data can be downloaded as needed.

Control group (isCGM)

Participants in the control group will continue using their own FreeStyle Libre to monitor glucose levels. Participants will have an access to FreeStyle Libre software and to the Tidepool website [33], which they can use at home to upload their readers and be able to review their blood glucose data and sensor data if they wish. No additional intervention will be provided during the 6-weeks isCGM period.

Intervention group (MM-CGM)

In addition to standard isCGM, the intervention group will receive the MM, which is a small Bluetooth transmitter cap that attaches to the isCGM sensor and is secured with a strap or a holder. The MM-CGM transmits sensor data to an app that displays continuous glucose data and provides safety alarms. Depending on the compatibility of the parent’s and child’s phones, the app can be xDrip+ or Tomato. No additional intervention will be provided during each 6-week intervention period (other than appropriate MM-CGM training/technical support). As a safety precaution for this novel study, participants will be instructed to perform finger stick glucose level measurements to confirm their blood glucose level before therapeutic interventions or corrective action if hypo- or hyperglycaemic glucose levels or symptoms occur. Also, the participants will be asked to do calibration to get accurate glucose readings from MM-CGM at least once per day, preferably in the morning before breakfast.

During the study, the set-up process and the training on how to use MM-CGM for all participants will be done by ME. The training will be done either face-to-face or online via Zoom. Participants in the MM-CGM phase will be provided with technical support as needed. Standard diabetes care will continue to be provided to participants by their usual paediatric diabetes care providers. Between scheduled study visits, participants will be referred back to their usual diabetes care providers for standard diabetes care, interpretation of readings, and optimisation of insulin dosages.

Wash-out phase

The wash-out phase will last for four weeks, and it will follow the first phase. The 4-week wash-out period is to eliminate any carry-over effect as there may be a chance to get accustomed to MM-CGM [26]. All participants during this phase will use isCGM to monitor their glucose levels. No further run-in week is required as a second baseline actigraph period will not be collected.

Procedures

After a 7-day run-in period, collection of the baseline measurements will be carried out. Study visits will be conducted by investigators or centrally trained delegates who have been trained on study procedures and data collection. There will be six visits for participants: visit 0, for assessing eligibility, collecting demographic data, and giving actigraphs to participants; visit 1, for returning actigraphs, filling in questionnaires, and randomisation; visit 2, one week before the end the first arm to wear actigraphs for all participants; visit 3, at the end of the first phase, returning actigraphs and filling in the questionnaires for all participants; visit 4, at the end of the wash-out phase to fill in questionnaires for all participants, and cross-over; visit 5, one week before the end of the second arm to wear actigraphs for all participants; visit 6, at the end of the second intervention and the end of the whole study, returning actigraphs and filling in questionnaires for all participants (Fig. 2).

At the end of the study, all participants will be able to keep their MM in addition to any other devices provided to them during the study; this includes smart phones used as main collecting or follower devices if needed. In addition, fuel vouchers will be provided as reimbursement for travelling/parking costs especially for those who live away from the hospital.

Outcome assessment

The schedule for assessing the primary outcome and the secondary outcomes is outlined in Table 3.

Table 3.

Schedule of outcomes assessment

	Screening, consent & randomisation		First arm		Wash-out period	Second arm
	One week		6 weeks		4 weeks	6 weeks
	Baseline Visit 0	Visit 1	Visit 2	Visit 3	Visit 4	Visit 5	Visit 6
	Week 0	Week 1	Week 5	Week 6	Week 10	Week 15	Week 16
Informed consent form	♣ ♦
Inclusion/Exclusion criteria	♣ ♦
Demographic	♣ ♦
HbA1c	♣ ♦
Height, Weight	♣ ♦
Actigraphs	♣ ♦		♣ ♦			♣ ♦
MiaoMiao education		♣			♦
HFSa		♣ ♦		♣ ♦			♣ ♦
Satisfaction
DTSQsb		♣ ♦			♣ ♦
DTSQcc				♣ ♦			♣ ♦
Quality of life
PedsQld generic core scales version 4.0		♣ ♦		♣ ♦	♣ ♦		♣ ♦
PedsQL diabetes module version 3.2		♣ ♦		♣ ♦	♣ ♦		♣ ♦
PedsQL family impact module version 2.0		♣ ♦		♣ ♦	♣ ♦		♣ ♦
PROMIS sleep questionnaires		♣ ♦		♣ ♦			♣ ♦
MiaoMiao user experience questionnaire				♣			♣

♣ MiaoMiao CGM + standard diabetes care

♦ isCGM + standard diabetes care

^aHypoglycaemia Fear Survey

^bDiabetes Treatment Satisfaction Questionnaire Parents/Guardian Baseline

^cDiabetes Treatment Satisfaction Questionnaire Parents/Guardian Change (6 weeks)

^dPediatric Quality of Life Inventory

Technical support and safety assessments

MM-CGM technical support

Technical support will be provided to the participants as needed alongside written guides designed for troubleshooting potential technical issues.

Safety assessments

Adverse events will be monitored at visit 3 and visit 6 for MM-CGM users only for hypoglycaemia, and loss of signal/loss of MM-CGM connectivity.

Participants will be referred to their usual emergency medical facilities for management of medical events such as severe hypoglycaemia or diabetic ketoacidosis.

Description of outcome measures

The hypoglycaemia fear survey (HFS)

The Hypoglycaemia Fear Survey (HFS) is a validated 25-item self-report measure of behaviours that children as well as their parents/caregivers may engage in as a result of FOH and specific worries related to various aspects of hypoglycaemia [30].

There are two subscales: Behaviour (10 items) and Worry (15 items). Scores are calculated as the overall mean of the items. Overall, higher scores reflect higher levels of FOH. Having higher score on the Behaviour Subscale shows that the respondent has a greater tendency to avoid hypoglycaemia and/or its negative consequences. The Worry Subscale reflects the level of worry concerning hypoglycaemia and its consequences; the higher the score, the more the worry the respondent has. Both HFS for parents and that for the children have the same subscale structure [34]. Only children aged 6–13 years, inclusive, will complete the HFS with the help from one of the research team as needed [34].

Anthropometry

Child participants will be asked to wear light clothing and remove their shoes before taking their height and weight measurements. Height will be measured to the nearest 0.1 cm using a standard stadiometer. Weight will be measured to the nearest 0.01 kg using a calibrated scale. BMI z-scores will be calculated via WHO standards using age and sex of the child [35].

Demographics

Demographic data (ie, date of birth, sex, ethnicity) will be collected from patients’ clinical records. The address where the participant lives more than 50% of the time will be used to assess their NZDep2013 score [36], a well-established marker of socioeconomic status in New Zealand. It is a deprivation index based on household address with one being the least deprived and ten being the most deprived [36].

Clinical data

Date of diabetes diagnosis will be used for subsequent calculation of duration of diabetes, insulin regimen, insulin dosing, HbA1c levels in the past three months, height and weight will be collected from recent clinical records.

Diabetes treatment satisfaction and diabetes technology questionnaire

The Diabetes Treatment Satisfaction Questionnaire-status version (DTSQs) is a validated 12-item self-report measure of a patient’s current treatment satisfaction. While the Diabetes Treatment Satisfaction Questionnaire- change version (DTSQc) measures the change in a patient’s satisfaction with their diabetes treatment regimen, which was developed to overcome potential ceiling effects (ie, where respondents score maximum or near-maximum satisfaction at baseline and can show little or no improvement at follow-up) [37].

A self-report 14-item parent version of the DTSQs measures parents’ satisfaction with their child diabetes treatment. The DTSQs will be used at baseline, and the DTSQc will be used at the end of each 6-week period to provide a ‘difference’ score for comparison.

The Diabetes Technology Questionnaire (DTQ) is a 30-item questionnaire which measures the impact of, and the degree of satisfaction with a technological tool which might be used in the management of T1D [38, 39]. The DTQ provides two separate scores one for the baseline or the current status, and the other one measures the change happened after usage. The higher scores represent a more favourable satisfaction and the impact rating of the technological toll which in this case will be MM-CGM.

Glucose monitoring device acceptability

Participants will self-report MM-CGM acceptability using a non-validated questionnaire. On a five point Likert scale: 1 (strongly disagree), 2 (Disagree), 3 (neutral), 4 (agree), and 5 (strongly agree). Participants will rate their opinion in regard to the following areas: acceptability of transmitter application, wear/use of the device, compatibility of pairing with isCGM and comparison to use of isCGM alone.

Quality of life

The Paediatric Quality of Life Inventory Generic Core Scales (PedsQLTM) is a validated 23-item self-report measure of children’s general health. Emotional, Social, and School Functioning Scales scores can be summed to provide a Psychosocial Health Summary Score. The physical Functioning Scale Score provides a Physical Health Summary Score. Higher scores indicate better health-related quality of life.

PedsQL version 4.0 Generic Core Scales will be used and it includes a parent proxy-report for ages 2–13 and, self-report for children ages between 5 and 13 years old inclusive [40]. The reliability, validity, and feasibility of the PedsQL has been demonstrated among children aged ≥ five years of age [41].

The Paediatric Quality of Life Inventory Diabetes Module for Type 1 diabetes (PedsQLTM Version 3.2) is a validated 28-item self-report measure of T1D specific health-related quality of life in [42] A self-report 28-item parent version measures their child T1D specific health-related quality of life. Higher scores indicate lower problems.

Parent quality of life will be investigated through PedsQL Family Impact Module (version 2.0). It is a validated self-reported 36 items parent report to measure the impact of the paediatric chronic conditions on the parent quality of life [43]. The PedsQL Family Impact Module has been validated for use among parents of children diagnosed with T1D [44].

Objectively measured sleep

Actigraphs

Children and one of their parents will wear an Actigraph (Actigraph, wGT3X-BT, Pensacola, FL, USA) continuously on the non-dominant wrist for up to seven days and eight nights, except during activities involving water (e.g., showering, swimming) at three-time points during the study: baseline; after five weeks of starting the first phase; and 5 weeks after starting the second phase. The participants will also complete a sleep diary over the same period to record when they go to sleep and wake up.

The data analysed will include sleep onset, offset, and night-time awakenings, to investigate dimensions of sleep health for example, sleep and wake timing, quality, quantity and variability. Actigraphs data will be adjusted for the number of weekend vs week days, and school holidays; daylight-saving periods will be avoided. Actigraphs will be initialized for 15 s epochs using ActiLife software (version 9.0.0). Data will be cleaned and scored using an automated script developed in MATLAB (MathWorks, Natick, MA, USA). It uses a count-scaled algorithm that automatically detects sleep onset and offset for overnight sleep and awakenings, after first inserting time-flags that approximate the child’s age [45]. The algorithm is validated for this device in children aged 8–12 years [46].

Sleep disturbance and sleep-related impairment

Patient-Reported Outcomes Measurement Information System (PROMIS) sleep questionnaires (short forms) will be used to evaluate sleep health in children and parents [47]. These include two sleep instruments that assess sleep disturbance and sleep impairment over the past seven days. The sleep disturbance instrument assesses sleep quality and continuity to provide information about concerns with getting to sleep, staying asleep, and adequacy and satisfaction with sleep. While, the sleep-related impairment instruments focus on measurement of waking alertness, sleeping, and tiredness during the usual waking hours of the day. We will be using the adult form for the parents sleep related disturbance and sleep related impairment. The parent proxy sleep disturbance form will be used for children aged 2 to 13 years, while parent proxy sleep related impairment forms for children aged 5–13 years will also be used. A self-reported measure for both sleep-related disturbance and sleep related impairment will be used for children 8–13 years old [47].

Glucose monitoring, levels, and trends

During each assessment, complete 2-week retrospective glucose readings will be downloaded using the Tidepool website [33]. This will be done for all study participants at baseline (visit0), at the final 2 weeks of each phase (visit 3 and visit 6), and at the final 2 weeks of the wash out period (visit 4). All the retrospective glucose readings collection points are shown in Table 1.

Sample size and power calculation

From previous literature, a conservative estimate of the SD of the parent-reported HFS is 20, and an estimate of the within-person correlation is 0.52 [48, 49]. In order to have 80% power to detect a difference of 8 points between the two treatments using a two-sided test at the 0.05 level, n = 50 participants with complete data for both treatments would be needed. This will be increased to 55 to allow for 10% missing data (e.g., data lost to periods of the child not wearing a sensor).

Data management

Data collection

Participants will be allocated a unique study identification (ID) number. Data will be stored electronically using a secure online platform provided by the University of Otago: REDCap [50].

Statistical analysis

The primary outcome is parental FOH and the primary analysis will follow the intent-to-treat principle with all participants analysed by the treatment to which they were randomised, regardless of actual MM-CGM/isCGM wear. To determine the difference in continuous outcomes (including the primary outcome of hypoglycaemia fear survey [HFS] scores) between treatments, linear mixed models will be used, adjusted for randomised order and with participant as a random effect. The centres will also be modelled using a random effect, although we do not anticipate this having any effect on results. The presence of carry-over and period effects will be explored by describing differences between treatments according to randomised order and by comparison to baseline measures. Two-sided p < 0.05 will be considered statistically significant. Missing data is unlikely to be substantial given our previous research but we have allowed for just under 10% loss to attrition in designing the study. Reporting of results will adhere to the Consolidated Standards of Reporting Trials (CONSORT) statement: extension to randomised cross-over trials [51].

Health and disability ethics committee review and informed consent

This study protocol was approved by the Northern Health and Disability Ethics Committee (19/NTB/118). The study was registered with the Australian New Zealand Clinical Trials Registry (ACTRN 12619001551189). Consultation with Māori (indigenous New Zealanders) was carried out through Ngāi Tahu Research Consultation Committee; input from the committee was considered by the researchers. The protocol underwent internal review by the research committee in the Women’s and Children’s Health Department, Dunedin School of Medicine, University of Otago. All participating district health boards approved the study recruitment and procedures. Informed consent will be obtained from all participating parents/caregivers while assent form will be obtained from children 7–13 years old. Parents/caregivers of children younger than seven years old will consent for their children.

Discussion

The management of T1D in children carries a significant psychological burden for both parents/caregivers and children; this burden is exaggerated by the FOH associated with intensive management of T1D. Currently, isCGM is indicated for reducing hypoglycaemia and potentially improving glycaemic control [52, 53]. However, CGM, a more expensive technology appears superior in these regards. With the introduction of DIY-CGM, many families have started to adopt this technology worldwide which also has the potential to reduce psychological burden and improve glycaemic control [26].

To our knowledge, this research is the first randomised study worldwide to elucidate the impact of DIY-CGM diabetes technology including the MiaoMiao plus isCGM (MiaoMiao CGM) in children. This research will provide much-needed evidence for (or against) clinicians recommending using open-source CGM in real-world settings. The randomised controlled cross-over design is a strength of the study, enabling comparisons with participants acting as their own control. Moreover, the current study will compare a DIY-CGM system to the very commonly used isCGM, which will also contribute to discussion around the utility of isCGM vs CGM in younger children. Finally, the outcomes of the current study span both core psychosocial and glycaemic aspects of diabetes care as well as sleep which will provide data regarding DIY-CGM device effectiveness in improving various core aspects of T1D management in children. Overall, it is hoped this data will contribute to the formulation of evidence-based advice to families impacted by T1D, and the healthcare workers caring for them, regarding the feasibility, safety and benefits of DIY-CGM.

Abbreviations

T1D

type 1 diabetes

DIY

do-it-yourself

FOH

Fear Of Hypoglycaemia

MM

MiaoMiao

CGM

Continuous Glucose Monitor

isCGM

Intermittently scanned Continuous Glucose Monitoring

MM-CGM

MiaoMiao Continuous Glucose Monitoring

DTSQs

Diabetes Treatment Satisfaction Questionnaire status

HFS

Hypoglycaemia Fear Survey

PROMIS

Patient-Reported Outcomes Measurement Information System

QoL

Quality of Life

REDCap

Research Data Capture

WHO

World Health Organisation

Authors’ contributions

BW is the grant holder and the principal investigator. All authors contributed to the study design. ME will coordinate technical aspects of the study protocol. ME, and SJ will conduct all recruitment and study visits. JH designed the statistical analysis plan. All authors contributed to refinement of the study protocol. ME drafted this manuscript. All authors contributed to, and approved the final manuscript.

Funding

This study is funded by a Lottery Health Research grant; and by the Department of Women’s and Children’s Health, Dunedin School of Medicine, University of Otago, New Zealand. Mona’s PhD funding is provided by a Freemasons Postgraduate Fellowship in Child Health, and a Freemason John Dennison Child Health Fellowship.

Data availability

Not applicable.

Compliance with ethical standards

Conflict-of-interest

The authors declare that they have no conflict of interest.

Ethics approval

This study has been approved by the Northern Health and Disability Ethics Committee (19/NTB/118).

Consent to participate

Consent to participate in this study will be obtained from all future participants as per the requirements of ISO 14155:2011 and Good Clinical Practices.

Consent for publication

Not applicable.

Code availability

Not applicable.