Predictive Factors of the Adherence to Real-Time Continuous Glucose Monitoring Sensors: A Prospective Observational Study (PARCS STUDY)

Abstract

Background:

Information about factors related to better adherence to continuous glucose monitoring (CGM) sensor adherence is quite limited.

Materials and Methods:

Forty-six participants with type 1 diabetes using continuous subcutaneous insulin infusion (CSII) without CGM were recruited. The participants’ characteristics and diabetes-related quality of life (QOL) were evaluated at baseline and one year after starting to use CGM. Participants wearing the sensor for ≥60% of the time were considered as adherent.

Results:

The mean age of the 46 participants was 44.1 ± 15.0 years old and the mean glycohemoglobin (HbA1c) was 7.7 ± 1.0%; 60.9% of the participants were classified as adherent. The duration of using CSII was longer in the adherent group, and the degree of diabetic retinopathy was significantly different. There were no significant differences in age, frequency of self-monitoring of blood glucose, or Hypoglycemia Fear Survey (HFS-B for behavior, HFS-W for worry) score at baseline between the adherent and nonadherent groups. The Problem Areas in Diabetes (PAID) score at baseline was significantly higher and the total CSII-QOL score at baseline was significantly lower in the adherent group. The usage of dual-wave bolus was significantly increased in the adherent group (34.6%-61.5%, P = .016), but not in the nonadherent group (33.3%-33.3%, P > .999). The HbA1c level showed a significant improvement in the adherent group (7.8%-7.3%, P < .001), but not in the nonadherent group (7.5%-7.2%, P = .102).

Conclusions:

Higher adherence to CGM sensors may be associated with a heavier emotional burden of diabetes and a worse QOL in relation to CSII at baseline.

Keywords

adherence CGM CSII HFS PAID QOL

Introduction

Type 1 diabetes (T1D) is a chronic condition that requires insulin therapy, leading to the potential adverse effect of hypoglycemia. Self-monitoring of blood glucose (SMBG) has been the standard method of glucose monitoring since the Diabetes Control and Complication Trial.¹ Indeed, the frequency of SMBG is negatively correlated with glycohemoglobin (HbA1c) in T1D patients.² Continuous glucose monitoring (CGM) was developed to overcome the lack of continuous information in SMBG.³ Previous studies, including the Juvenile Diabetes Research Foundation (JDRF) and STAR-3 studies, demonstrated a relationship between HbA1c improvement and better CGM sensor adherence.^4-8 However, information about the factors related to the better adherence to CGM sensors is quite limited. For example, in the JDRF study, adherence to CGM sensors was associated with age ≥25 years and more frequent self-reported blood glucose meter measurements per day; other factors were unclear.⁷ In this study, improvement in quality of life (QOL) in relation to fear of hypoglycemia was observed.⁸ As self-management behavior may be generally affected by psychosocial factors, we hypothesized that QOL related to factors such as fear of hypoglycemia and the psychological burden of diabetes, income, or need to drive might be associated with a higher frequency of CGM sensor usage.

To investigate factors related to better adherence to CGM sensors, we performed a prospective observational study of participants with T1D who were using continuous subcutaneous insulin infusion (CSII) and who started real-time CGM for the first time.

Materials and Methods

This study was an open-label, single-arm, multicenter observational study conducted at nine institutes in Japan (National Hospital Organization Kyoto Medical Center, Tokushima University, Tokai University School of Medicine, Kobe University Graduate School of Medicine, National Hospital Organization Osaka National Hospital, Arisawa General Hospital, Okayama University Hospital, Kanda Naika Clinic, and Okada Clinic). Forty-six T1D participants using insulin pumps without CGM were recruited between February 2015 and January 2017. The participants were followed up for one year after switching to the MiniMed 620G (Medtronic, Inc., Northridge, USA) sensor-augmented pump (SAP). The study was approved by the Ethical Committee of NHO Kyoto Medical Center (14-088) and was registered in the UMIN Clinical Trials Registry (UMIN-CTR: UMIN000016588). Written informed consent was obtained from all participants prior to their participation in the study.

Study Population and Survey

The inclusion and exclusion criteria are described in Table 1. Information regarding the following baseline characteristics was collected: age, sex, duration of diabetes, duration of CSII use, model of insulin pump used prior to the study, diabetic retinopathy (none, background retinopathy, pre-proliferative retinopathy, proliferative retinopathy, or post-photocoagulation), diabetic nephropathy (none, microalbuminuria, macroalbuminuria, renal failure, or end-stage renal disease), HbA1c, body mass index (BMI), blood pressure (systolic, SBP and diastolic, DBP), total daily insulin dose (TDD), total daily basal insulin dose (TBD), frequency of SMBG, coverage by social security, employment (full-time employment, part-time employment, student, homemaker, pensioner, or unemployed), possession of driver’s license, driving mileage, and annual income. HbA1c was measured at each medical institute satisfying the standards determined by JDS.⁹ Severe adverse events (SAEs) were determined as death or hospitalization due to either severe hypoglycemia or diabetic ketoacidosis. Severe hypoglycemia was determined as hypoglycemia for which treatment required assistance.¹ Multiple episodes of adverse events (AEs) in the same participant were counted as one incidence.

Table 1.

Inclusion and Exclusion Criteria.

Inclusion criteria
Age ≥15 y
T1D
Use of an insulin pump for ≥1 y
Participants willing to use SAP
Participants who regularly visit or were hospitalized at the medical institute participating in this study

Exclusion criteria
Previous usage of real-time CGM or SAP
Pregnancy or undergoing preconception care
Age <15 y
Participants not suitable to participate in this study for other reasons

Abbreviations: CGM, continuous glucose monitoring; SAP, sensor-augmented pump; T1D, type 1 diabetes.

A diagnosis of T1D was made according to the criteria of the Japan Diabetes Society (JDS).¹⁰

Psychometric Evaluation

To assess the QOL, the Hypoglycemia Fear Survey (HFS; HFS-B for behavior, HFS-W for worry) was used to evaluate the fear of hypoglycemia, Problem Areas in Diabetes (PAID; 20 items) was used to evaluate the emotional burden of diabetes, the Euro-QOL 5 dimensions 3-level version (EQ-5D-3L) utility score was used to measure the functional health status, and CSII-QOL (convenience factor, social restriction factor, and psychological problems factor) was used to evaluate the QOL in relation to CSII.^11-14 A higher HFS score indicates worse fear of hypoglycemia, a higher PAID score indicates a heavier emotional burden of diabetes, a lower EQ-5D-3L score indicates a poorer functional health status, and a lower CSII-QOL score indicates worse CSII-related QOL. All of the psychometric evaluations were conducted in Japanese. The Japanese versions of PAID¹⁵ and EQ-5D-3L were linguistically validated, and the validated CSII-QOL was originally developed in Japanese. However, the Japanese version of HFS (B&W) is not back-translated despite it having been widely used for more than a decade.¹⁶

Statistical Analyses

CGM data were downloaded to PCs for the analysis. Sensor adherence was calculated as the time using the sensor per week using CareLink Pro Therapy Management Software Ver. 4.0C (Medtronic, Inc., Northridge, USA). Based on the observation in the STAR-3 study, participants wearing the sensor for ≥60% were considered adherent⁵. The adjusted mean change in body weight and HbA1c at year 1 was obtained using the last observation carried forward. Missing data in the QOL survey were omitted. For continuous variables, the Student’s t-test was used to compare the mean values of the two study arms. For categorical variables, the X²-test or Fisher’s exact test was used to compare proportions. Pearson’s correlation efficient was calculated to assess correlation. Spearman’s correlation efficient was calculated to assess correlation with ranking. P values of <.05 were considered to indicate statistical significance. Data are expressed as the mean (standard deviation) or percentage. All statistical analyses were conducted using SPSS for Windows Ver. 23.0 (IBM, Armonk, NY, USA) or EZR.¹⁷

Results

Forty-six participants were recruited, and 45 completed the study. One participant withdrew six months after the start of the study.

The baseline characteristics of the participants are described in Table 2. Twenty-eight participants (60.9%) were classified into the adherent group and 18 (39.1%) were classified into the nonadherent group. The overall time wearing the sensor was 61.9 ± 26.8%. In the adherent group, the time wearing the sensor was significantly longer in comparison to the nonadherent group (80.1 ± 10.8% vs 33.5 ± 18.1%, P < .001).

Table 2.

The Baseline Characteristics of the Adherent and Nonadherent Groups.

Variables	Total (n = 46)	Adherent (n = 28)	Nonadherent (n = 18)	P value
Age, y	44.1 (15.0)	45.8 (14.5)	41.5 (15.9)	.356
Female, %	73.9	67.9	83.3	.243
Diabetes duration, y	18.8 (13.6)	19.1 (11.7)	18.3 (16.5)	.849
CSII, y	3.7 (2.4)	4.3 (2.3)	2.7 (2.4)	.028^*
Model of insulin pump
Paradigm 712	2	2	0	.231
Paradigm 722	30	20	10
MiniMed 620G	14	6	8
Retinopathy
None	30	15	15	.029^*
Background retinopathy	9	7	2
Pre-proliferative retinopathy	1	0	1
Proliferative retinopathy	0	0	0
Post-photocoagulation	6	6	0
Nephropathy
None	40	24	16	.354
Microalbuminuria	2	1	1
Macroalbuminuria	3	3	0
Renal failure	1	0	1
ESRD	0	0	0
HbA1c, %	7.7 (1.0)	7.8 (0.9)	7.5 (1.1)	.270
BMI, kg/m²	22.7 (2.8)	22.7 (2.6)	22.8 (3.2)	.858
SBP, mmHg	118 (13)	117 (14)	120 (13)	.440
DBP, mmHg	72 (10)	71 (10)	73 (12)	.605
TDD, units/day	36.1 (13.8)	37.3 (12.8)	34.3 (15.4)	.483
TBD/TDD, %	33.8 (12.4)	34.8 (13.4)	32.2 (11.0)	.493
Frequency of SMBG, times/day	4.8 (1.8)	5.1 (1.9)	4.5 (1.5)	.299
Coverage by social security, %	8.7	7.1	11.1	.639
Status of employment
Full-time employment	16	10	6	.791
Part-time employment	11	7	4
Student	3	2	1
Homemaker	9	6	3
Pensioner	2	1	1
Unemployed	4	1	3
Possession of driving license
Yes	38	23	15	>.999
No	7	4	3
Driving mileage, km/y
≤3000	20	12	8	.648
3001-5000	10	5	5
5001-10 000	3	2	1
10 001-15 000	2	0	2
≥15 000	2	1	1
Income (million yen/y), %
<2	6	4	2	.743
2-6	27	15	12
≥6	10	7	3
History of severe hypoglycemia within one year, %	13.3	11.5	16.7	.676
Frequency of self-reported hypoglycemia (<70 mg/dL) per month	7.0 (4.0, 13.5)	5.0 (4.0, 15.0)	8.5 (4.0, 10.0)	.860

Abbreviations: BMI, body mass index; CSII, continuous subcutaneous insulin infusion; DBP, diastolic blood pressure; ESRD, end-stage renal disease; HbA1c, glycated hemoglobin; SBP, systolic blood pressure; SD, standard deviation; SMBG, self-monitoring of blood glucose; TBD, total daily basal insulin dose; TDD, total daily insulin dose.

Data are presented as the number, percentage or mean (SD), median (25%, 75%), as indicated.

P < .05.

The duration of using CSII use was longer in the adherent group, and the degree of diabetic retinopathy was significantly different in the adherent group (Table 2). The ratio of post-photocoagulation diabetic retinopathy in the adherent and nonadherent groups was not significantly different (13.0% vs 0.0%, P = .068). After one year, there was a significant increase of BMI in the overall study population and in the adherent group (Table 3). There was also a significant improvement in HbA1c in the overall study population and in the adherent group, but not in the nonadherent group. There was a significant increase of TDD in the overall study population and in the nonadherent group.

Table 3.

Change in Characteristics After One Year.

Variables	Total		Adherent		Nonadherent		P value
Variables	n	Mean (SD)	n	Mean (SD)	n	Mean (SD)	P value
BMI
Baseline	46	22.7 (2.8)	28	22.7 (2.6)	18	22.8 (3.2)	.858
1-y	46	23.2 (3.1)	28	23.3 (2.9)	18	23.0 (3.6)	.794
P value	.010^*		.003^*		.548
SBP, mmHg
Baseline	45	118.4 (13.5)	28	117.0 (13.7)	17	120.8 (13.2)	.440
1-y	45	117.8 (12.9)	28	114.6 (11.2)	17	123.1 (14.1)	.031
P value	.700		.195		.452
DBP, mmHg
Baseline	45	71.9 (10.1)	28	70.9 (9.5)	17	73.6 (11)	.605
1-y	45	71.5 (10.5)	28	70.4 (9.6)	17	73.4 (11.8)	.369
P value	.740		.716		.920
HbA1c, %
Baseline	46	7.7 (1.0)	28	7.8 (0.9)	18	7.5 (1.1)	.270
1-y	46	7.3 (0.7)	28	7.3 (0.7)	18	7.2 (0.7)	.571
P value	<.001^*		.001^*		.102
TDD
Baseline	46	36.1 (13.8)	28	37.3 (12.8)	18	34.3 (15.4)	.483
1-y	46	39.5 (14.2)	28	39.4 (13.4)	18	39.7 (15.7)	.932
P value	.023^*		.28		.023^*
TBD/TDD, %
Baseline	46	33.8 (0.1)	28	34.9 (0.1)	18	32.2 (0.1)	.493
1-y	46	32.9 (0.1)	28	33.2 (0.1)	18	32.5 (0.1)	.823
P value	.482		.265		.935
Frequency of SMBG, times/day
Baseline	46	4.8 (1.8)	46	5.1 (1.9)	46	4.5 (1.5)	.299
1-y	46	4.7 (1.5)	46	4.8 (1.5)	46	4.5 (1.7)	.544
P value	.464		.341		>.999
Frequency of bolus
Baseline	40	5.0 (1.8)	24	4.8 (1.9)	16	5.2 (1.7)	.515
1-y	40	5.3 (2.8)	24	5.2 (1.5)	16	5.5 (4.0)	.592
P value	.368		.268		.723
Frequency of self-reported hypoglycemia (symptomatic or <70 mg/dL) per month
Baseline	7.0	(4.0, 15.0)	6.0	(4.0, 15.0)	10.0	(4.0, 10.0)	.924
1-y	8.0	(5.0, 15.0)	10.0	(2.6, 15.0)	6.0	(5.0, 12.5)	.786
P value	.279		.476		.396

Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; HbA1c, glycated hemoglobin; SBP, systolic blood pressure; SD, standard deviation; SMBG, self-monitoring of blood glucose; TBD, total daily basal insulin dose; TDD, total daily insulin dose.

Data are presented as the number, percentage or mean (SD), median (25%, 75%), as indicated.

P < .05.

There was no significant difference in HFS (total HFS score, HFS-B score, and HFS-W score) at baseline between the adherent and nonadherent groups (Table 4). There was a significant improvement in the HFS-W score after one year in the overall study population. In the adherent group, the PAID score at baseline was significantly higher in comparison to the nonadherent group. There was no significant change in the PAID score after one year. There was a significant worsening in the EQ-5D-3L utility score after one year in the nonadherent group.

Table 4.

HFS, PAID, and EQ-5D Utility Score.

Variables	Total		Adherent		Nonadherent		P value
Variables	n	Mean (SD)	n	Mean (SD)	n	Mean (SD)	P value
HFS (total)
Baseline	43	34.3 (16.8)	26	37.5 (17.5)	17	29.4 (14.7)	.112
1-y	43	30.3 (13.7)	26	31.9 (14.2)	17	28.0 (12.9)	.218
P value	.057		.064		.584
HFS-B
Baseline	43	16.9 (6.6)	26	18.2 (6.2)	17	15.0 (6.8)	.101
1-y	43	16.9 (6.8)	26	16.9 (6.7)	17	16.8 (7.0)	.961
P value	.952		.192		.15
HFS-W
Baseline	43	17.3 (12.0)	26	19.3 (13.1)	17	14.4 (9.6)	.178
1-y	43	13.5 (8.9)	26	15.0 (9.5)	17	11.2 (7.6)	.165
P value	.024^*		.093		.109
PAID
Baseline	44	34.7 (17.0)	27	38.1 (18.2)	17	29.4 (13.8)	.044^*
1-y	44	33.5 (17.2)	27	36.0 (16.0)	17	29.5 (18.8)	.227
P value	.531		.387		.984
EQ-5D-3L utility
Baseline	43	0.897 (0.159)	26	0.876 (0.165)	17	0.928 (0.147)	.412
1-y	43	0.875 (0.159)	26	0.902 (0.157)	17	0.835 (0.158)	.184
P value	.339		.360		.007^*

Abbreviations: EQ-5D-3L, Euro-QOL 5 dimensions 3-level version; HFS, Hypoglycemia Fear Survey (HFS-B for behavior, HFS-W for worry); PAID, Problem Areas in Diabetes; SD, standard deviation.

Data are presented as the number or mean (SD).

P < .05.

Each item of the PAID self-reported questionnaire was compared between the adherent and nonadherent groups, and the scores for items “7. Not knowing if your mood or feelings are related to your diabetes?”, “11. Feeling constantly concerned about food and eating?”, “13. Feelings of guilt or anxiety when you get off track with your diabetes management?”, “19. Coping with complications of diabetes?”, and “20. Feeling “burned out” by the constant effort needed to manage diabetes?” were significantly higher in the adherent group (Supplemental Table S1). Answers of severe distress (PAID scores 3 and 4) for items “7” and “13” were significantly frequent in the adherent group (Supplemental Table S2). There was no significant correlation between the duration of using CSII and the PAID score (r = –0.004, P = .981). The degree of diabetic retinopathy was not correlated with the PAID score (Supplemental Table S3).

In the adherent group, the total CSII-QOL score at baseline and the score for psychological problems at baseline were significantly lower in comparison to the nonadherent group (Table 5). After one year, the score for convenience significantly improved in comparison to that at baseline in the overall study population, and in the adherent and nonadherent groups.

Table 5.

CSII-QOL Score.

Variables	Total		Adherent		Nonadherent		P value
Variables	n	Mean (SD)	n	Mean (SD)	n	Mean (SD)	P value
Total score
Baseline	44	60.1 (14.3)	27	57.0 (13.9)	17	65.1 (13.8)	.045^*
1-y	44	63.4(16.4)	27	61.9 (16.9)	17	65.9 (15.8)	.428
P value	.090		.073		.763
Convenience
Baseline	44	16.3 (2.3)	27	16.5 (2.3)	17	15.9 (2.3)	.486
1-y	44	20.0 (3.6)	27	20.5 (3.2)	17	19.1 (4.1)	.223
P value	<.001^*		<.001^*		.003^*
Social restriction
Baseline	44	25.0 (6.6)	27	23.6 (6.5)	17	27.2 (6.2)	.057
1-y	44	24.8 (7.3)	27	23.8 (7.9)	17	26.2 (6.3)	.291
P value	.786		.889		.422
Psychological problems
Baseline	44	18.8 (7.6)	27	16.8 (7.5)	17	22.1 (7.0)	.022^*
1-y	44	18.7 (8.0)	27	17.6 (8.4)	17	20.6 (7.2)	.225
P value	.911		.613		.241

Abbreviations: CSII, continuous subcutaneous insulin infusion; QOL, quality of life; SD, standard deviation.

Data are presented as the mean (SD) or number.

P < .05.

The usage of dual-wave bolus increased significantly in the adherent group but not in the nonadherent group (Supplemental Table S4).

No SAEs were observed in either groups. Episodes of severe hypoglycemia that did not result in hospitalization were observed in two participants in the adherent group (two episodes each). No episodes of diabetic ketoacidosis that did not result in hospitalization were observed. Contact dermatitis due to adhesive was observed in one participant each in the adherent and nonadherent groups. The incidence was 12.8 cases/100 person-years for total AEs, 8.5 cases/100 person-years for AEs related to the study, 4.3 cases/100 person-years for severe hypoglycemia, and 4.2 cases/100 person-years for contact dermatitis due to adhesive. There were no statistically significant differences between the adherent and nonadherent groups in the incidence total AEs (17.8 vs 5.2 cases/100 person-years, P = .217), AEs related to the study (10.6 vs 5.2 cases/100 person-years, P = .519), severe hypoglycemia (7.2 vs 0.0 cases/100 person-years, P = .256), or contact dermatitis due to adhesive (3.5 vs 5.2 cases/100 person-years, P = .789).

Discussion

We performed a prospective observational study to investigate factors associated with adherence to CGM in T1D participants using insulin pumps. As there was a domestic issue that a standalone CGM device separate from an insulin pump was not commercially available in Japan when the current study was initiated, only patients who were already using a CSII were recruited. In addition, the incidence of T1D in East Asian populations is much lower than that in Caucasian populations¹⁸; thus, there was difficulty in recruiting participants. As a result, only 46 participants were recruited from nine institutes. The majority of participants in this study were female, which was consistent with the observation in a previous study regarding the usage of CSII among T1D patients in Japan.¹⁹ The reason why a longer duration of CSII use was associated with better adherence to CGM remains unclear. The reason why there was a significant difference in the degree of diabetic retinopathy between the two groups also remains unclear; however, it is possible that participants with advanced diabetic retinopathy might want to use CGM more constantly to prevent a further worsening of the complication. The reason why age was not associated with CGM adherence may be due to the higher age of the T1D participants in this study. The frequency of SMBG was positively associated with adherence to CGM in the JDRF study,⁷ but not in this study. One of the largest differences between the JDRF study and the current study is that the participants in the JDRF study included both patients using multiple daily injections (MDI) and CSII; the present study only included participants using CSII.

The significant improvement of HbA1c in the overall study population (mean time wearing the sensor: 61.9%) and in the adherent group (mean time wearing the sensor: 80.1%), but not in the nonadherent group (mean time wearing the sensor: 33.5%) was consistent with previous studies.^4,5 The reason why there was a significant increase in the BMI in the overall study population and in the adherent group, but not in the nonadherent group, remains unclear. The reason why there was significant increase in the TDD in the overall study population and nonadherent group, but not in the adherent group, also remains unclear.

We found that a higher baseline PAID score was associated with better adherence to CGM usage. The baseline HFS score (total, behavior, and worry) and EQ-5D-3L utility score were not associated with adherence to CGM in this study. The reasons for these observations are not clear; however, it is possible that participants who felt a heavy burden of diabetes even after starting CSII may use CGM more frequently. It is also possible that participants with better adherence to CGM may include individuals who feel a heavier burden of diabetes. Indeed, there were significant differences between the adherent and nonadherent groups in scores for PAID items related to difficulty in coping with T1D.

The absence of improvement in the PAID score after one year of CGM is of interest, as it suggests that CGM did not significantly relieve the burden of diabetes in this study population. The current study might have failed to detect the change in the PAID score due to the small sample size, or the effect of real-time CGM on QOL could differ depending on the patient background or the modalities used to evaluate QOL. The DIAMOND study conducted in the United States, which exclusively recruited patients using MDI reported an improvement in Diabetes Distress Scale.²⁰ The RESCUE Trial conducted in Belgium, which exclusively recruited patients using CSII, reported an improvement in the Problem Area in Diabetes-short form and Short Form 36.²¹ These studies did not analyze the association between diabetes-related QOL and CGM adherence.

In contrast, the HFS-W score significantly improved in the overall study population after one year. This observation was similar to the RESCUE Trial,²¹ but differed from the JDRF study in which the HFS-B score improved after CGM usage⁸ or from the DIAMOND and Hypo-DE studies in which there was no significant change in the HFS-W score.^20,22 The reason why there was a significant worsening in the EQ-5D-3L utility score after one year in the nonadherent group remains unclear. There was no significant difference in the EQ-5D Utility Index between the rt-CGM arm and control arm in Hypo-DE study.²² One possibility is that there might be some confounding factors between the worsening of the EQ-5D-3L utility score and the nonadherence to CGM.

We also used the recently developed CSII-QOL score, which was created to assess the QOL specifically related to CSII.¹⁴ Better adherence to CGM was associated with worse CSII-related QOL, especially with psychological problems, which suggests that participants who had unmet needs in the self-management of T1D, even after starting CSII, might have wanted to perform CGM more constantly.

The observation that the usage of dual-wave bolus increased significantly in the overall study population and in the adherent group, but not in the nonadherent group, may suggest an important role of CGM in the management of postprandial hyperglycemia. Watching daily glucose excursion through CGM might have motivated the participants to use dual-wave bolus to manage the prolonged postprandial hyperglycemia accompanying the consumption of fat and protein rich food.

The incidence of severe hypoglycemia (4.3 cases/100 person-years) in this study was low; however, this rate might not be accurate due to the relatively small study population.

Although not significantly different, the incidence of severe hypoglycemia in the adherent group tended to be higher than that in the nonadherent group. As real-time CGM is reported to be useful for reducing hypoglycemic events in T1D patients using MDI,²² it is possible that patients with a higher risk of developing severe hypoglycemia might have wanted to use CGM more frequently, but the sample size and duration of the current study were not sufficient to compare the incidence of severe hypoglycemia in both groups. A future study with a larger population and longer duration will be necessary to address this issue, combined with a survey of the past history of severe hypoglycemia at baseline.

The present study was associated with some limitations. First, the study population was relatively small. Second, it employed an open-label, single-arm observational design. Third, only T1D participants using an insulin pump were included and those using MDI were excluded, as no real-time CGM separate from the insulin pump was available in Japan when this study was planned. Fourth, factors that might underlie poor adherence, such as dissatisfaction with sensor accuracy and pain, were not investigated in this study.

Conclusion

In this study, a heavier psychological burden of diabetes, as measured by the PAID score and a worse CSII-related QOL, as measured by the CSII-QOL score before the start of CGM, were associated with better adherence to CGM (≥60%) after its introduction in participants with T1D who were already using CSII. The PAID score was not improved after CGM, but worry in relation to hypoglycemia, as measured by the HFS-W score, improved after CGM. Our results suggest that the assessment of psychological status in T1D patients before starting CGM may predict adherence to CGM after its introduction.

Supplemental Material

Murata_T_PARCS_study_Supplementary_File_rev2 – Supplemental material for Predictive Factors of the Adherence to Real-Time Continuous Glucose Monitoring Sensors: A Prospective Observational Study (PARCS STUDY)

Supplemental material, Murata_T_PARCS_study_Supplementary_File_rev2 for Predictive Factors of the Adherence to Real-Time Continuous Glucose Monitoring Sensors: A Prospective Observational Study (PARCS STUDY) by Takashi Murata, Akio Kuroda, Munehide Matsuhisa, Masao Toyoda, Moritsugu Kimura, Yushi Hirota, Ken Kato, Hideaki Sawaki, Atsuhito Tone, Satoshi Kawashima, Akira Okada, Tomokazu Watanabe, Shinsuke Nirengi, Akiko Suganuma and Naoki Sakane in Journal of Diabetes Science and Technology

Footnotes

Acknowledgements

The authors thank Dr. Junnosuke Miura (Diabetes Center, Tokyo Women’s Medical University School of Medicine, Tokyo, Japan) for indispensable discussion and critical reading of the manuscript. Abstracts of the study in Japanese were presented at the 18th Association for the Study of Innovative Diabetes Treatment in Japan (ASINDTJ) annual meeting (September 29-30, 2018, Kobe, Japan) and 19th ASINDTJ annual meeting (November 2nd-3rd, 2019, Tokyo, Japan), and an abstract of the study in English was presented at the 12th Advanced Technologies and Treatments for Diabetes (ATTD) annual meeting (February 20-23, 2019, Berlin, Germany).

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: AK discloses the following relationships: lecture fees from Medtronic, Terumo, and Abbott. MM discloses the following relationships: lecture fees from Medtronic and Abbott. MT discloses the following relationships: lecture fees from Medtronic, Terumo, and Abbott. MK discloses the following relationships: lecture fees from Medtronic. YH discloses the following relationships: lecture fees from Medtronic, Terumo, and Abbott. KK discloses the following relationships: lecture fees and manuscript fees from Medtronic, and lecture fees from Terumo. AT discloses the following relationships: lecture fees from Medtronic, and lecture fees and manuscript fees from Abbott. SK discloses the following relationships: lecture fees from Medtronic and Terumo. The other authors declare no conflicts of interest in association with the present study.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by JSPS KAKENHI Grant Number 18K01988 and a grant from the National Hospital Organization (NHO) network study.

ORCID iD

Takashi Murata

Supplemental Material

Supplemental material for this article is available online.

References

The Diabetes Control and Complications Trial Research Group, Nathan

Genuth

, et al.The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329(14):977-986.

Murata

Tsuzaki

Yoshioka

, et al. The relationship between the frequency of self-monitoring of blood glucose and glycemic control in patients with type 1 diabetes mellitus on continuous subcutaneous insulin infusion or on multiple daily injections. J Diabetes Investig. 2015;6(6):687-691.

Rodbard

Continuous glucose monitoring: a review of successes, challenges, and opportunities. Diabetes Technol Ther. 2016;18(suppl 2):S3-S13.

Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, Tamborlane

Beck

, et al. Continuous glucose monitoring and intensive treatment of type 1 diabetes. N Engl J Med. 2008;359(14):1464-1476.

Bergenstal

Tamborlane

Ahmann

, et al. Effectiveness of sensor-augmented insulin-pump therapy in type 1 diabetes. N Engl J Med. 2010;363(4):311-320.

Battelino

Conget

Olsen

, et al. The use and efficacy of continuous glucose monitoring in type 1 diabetes treated with insulin pump therapy: a randomised controlled trial. Diabetologia. 2012;55(12):3155-3162.

Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, Beck

Buckingham

, et al. Factors predictive of use and of benefit from continuous glucose monitoring in type 1 diabetes. Diabetes Care. 2009;32(11):1947-1953.

Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group, Beck

Lawrence

, et al. Quality-of-life measures in children and adults with type 1 diabetes: juvenile diabetes research foundation continuous glucose monitoring randomized trial. Diabetes Care. 2010;33(10):2175-2177.

Kashiwagi

Kadowaki

Haneda

, et al. Consensus and statement on international standardization of HbA1C in Japan: committee report on diabetes mellitus laboratory testing standardization. J Jpn Diabetes Soc. 2009;52(9):811-818.

10.

Committee of the Japan Diabetes Society on the Diagnostic Criteria of Diabetes Mellitus, Seino

Nanjo

, et al. Report of the committee on the classification and diagnostic criteria of diabetes mellitus. J Diabetes Investig. 2010;1(5):212-228.

11.

Gonder-Frederick

Schmidt

Vajda

, et al. Psychometric properties of the hypoglycemia fear survey-ii for adults with type 1 diabetes. Diabetes Care. 2011;34(4):801-806.

12.

Welch

Jacobson

Polonsky

WH.

The problem areas in diabetes scale. An evaluation of its clinical utility. Diabetes Care. 1997;20(5):760-766.

13.

Shiroiwa

Fukuda

Ikeda

, et al. Japanese population norms for preference-based measures: EQ-5D-3L, EQ-5D-5L, and SF-6D. Qual Life Res. 2016;25(3):707-719.

14.

Sakane

Murata

Tone

, et al. Development and validation of the continuous subcutaneous insulin infusion-related quality of life (CSII-QOL) scale. Diabetes Technol Ther. 2020;22(3):216-221.

15.

Ishii

Welch

Jacobson

, et al. The Japanese version of problem area in diabetes scale: a clinical and research tool for the assessment of emotional functioning among diabetic patients(abstract). Diabetes. 1999;48(5):A319.

16.

Kitaoka

Controlling psycho-social factors. (in Japanese)

Chiryo. 2001;83:67-72.

17.

Kanda

Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant. 2013;48(3):452-458.

18.

Kawasaki

Matsuura

Eguchi

Type 1 diabetes in Japan. Diabetologia. 2006;49(5):828-836.

19.

Murata

Aoki

Kato

, et al. The percentage of continuous subcutaneous insulin infusion usage among adult type 1 diabetes mellitus patients in Japan: a cross-sectional study at national hospital organization hospitals. J Diabetes Sci Technol. 2017;11(5):1055-1056.

20.

Polonsky

Hessler

Ruedy

, et al. The impact of continuous glucose monitoring on markers of quality of life in adults with type 1 diabetes: further findings from the DIAMOND randomized clinical trial. Diabetes Care. 2017;40(6):736-741.

21.

Charleer

Mathieu

Nobels

, et al. Effect of continuous glucose monitoring on glycemic control, acute admissions, and quality of life: a real-world study. J Clin Endocrinol Metab. 2018;103(3):1224-1232.

22.

Heinemann

Freckmann

Ehrmann

, et al. Real-time continuous glucose monitoring in adults with type 1 diabetes and impaired hypoglycaemia awareness or severe hypoglycaemia treated with multiple daily insulin injections (HypoDE): a multicentre, randomised controlled trial. Lancet. 2018;391(10128):1367-1377.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.23 MB