Comparison of the Acceptance of Telemonitoring for Glucose Management Between South Korea and China

Abstract

Background:

Telemonitoring is used for glucose management and support in many countries. A better understanding of the differences in telemonitoring acceptance based on regional characteristics is needed. Therefore, we compared the acceptance of telemonitoring for glucose management among patients in South Korea and China.

Materials and Methods:

This study used data from Korean (n = 81) and Chinese (n = 92) subjects with type 2 diabetes. We used two independent sample t-tests to compare patients' perceptions of telemonitoring and multiple regression analysis to determine the factors that affected their behavioral intentions to use telemonitoring. We conducted Wilcoxon signed rank tests to assess the differences in hemoglobin A1c (HbA1c) levels from baseline to follow-up.

Results:

Although Korean and Chinese patients had positive perceptions of the services, different factors influenced their behavioral intentions to use them. In South Korea, performance expectations and social influences were significantly associated with intention to use telemonitoring. Patients younger than 50 years showed a significant decrease in HbA1c levels at month 6 (p < 0.05). In China, effort expectancy and facilitating conditions were significantly associated with intention to use. In addition, subjects in all age groups exhibited a significant decrease in HbA1c levels at all follow-up points (p < 0.001).

Conclusion:

Telemonitoring was a supportive intervention in improving blood sugar levels among patients with diabetes in South Korea and China, but the factors influencing its use varied. We provide practical guidance for developing telemonitoring for glucose management that considers the distinct characteristics of different countries.

Introduction

Continuous care for diabetes requires management of various factors, including diet, exercise, medications, and blood glucose levels. Diabetes requires periodic assistance from a medical provider over a long period of time. A crucial component of treating diabetes is controlling blood sugar and preventing diabetes complications. Telemonitoring provides an advantage in diabetes management because it enables constant support for glucose management.

Telemonitoring is a useful intervention for managing blood sugar regardless of where patients with diabetes live. Telemonitoring is used for supporting patients in their diabetes management in many countries.^1

–9 In South Korea and China specifically, randomized controlled trials have been conducted to demonstrate the importance of remote management of type 2 diabetes in reducing the progression of diabetes complications.^10,11 Although this technology exists in a similar form worldwide, patients with diabetes behavior and their acceptance of technology may vary by country. To develop telemonitoring services that account for regional differences, there is a need to first understand the telemonitoring service acceptance and users' characteristics. Accordingly, we focused on technology acceptance in South Korea and China.

Although telemonitoring services can support patients' management of blood sugar levels and diabetes complications, they are only valuable when used properly and when patients accept the technology. Patient acceptance of a new technology has a profound influence on its success. Thus, telemonitoring services may have issues related to patient acceptance if the patients are unfamiliar with the technology used. Much of the healthcare research related to technology acceptance has focused on models such as the technology acceptance model (TAM) and the unified theory of acceptance and use of technology (UTAUT).^{10,12

–18} These studies used the constructs of performance expectations, effort expectancy, social influence, and facilitation conditions. These constructs provide a comprehensive understanding of the acceptance of technology among patients.

Based on this background, this study was conducted to examine the acceptance of telemonitoring for glucose control in South Korea and China. To gain deeper insight into telemonitoring services while considering regional characteristics, we used the key constructs of the UTAUT.

Research Design and Methods

Instrument development

The UTAUT includes four key constructs: performance expectancy, effort expectancy, social influence, and facilitation conditions. This study's main variables were based on the UTAUT model. These constructs provide a comprehensive framework for user technology acceptance. This study included six final constructs: (1) performance expectancy, (2) effort expectancy, (3) social influence, (4) facilitation conditions, (5) behavioral intention to use, and (6) satisfaction, listed in Table 1.

Table 1.

Constructs and Questionnaire Items

CONSTRUCT AND MEASUREMENT ITEMS	REFERENCE
Performance expectancy	19, 20, 21
I am able to monitor my blood glucose.
The telemonitoring service improves my blood glucose management.
The service helps me self-manage my blood glucose levels.
Effort expectancy	20
I find this service easy for taking my blood pressure, checking my blood sugar level, and saving the results.
My interaction with the system is clear and understandable.
It would be easy for me to become skillful at using the system.
Social influence	20
People who influence my healthcare think that I should use the telemonitoring service.
People who are important to me think that I should use the telemonitoring service.
People who have a disease similar to mine should use the telemonitoring service.
Facilitating conditions	20, 22
I am able to receive assistance and guidelines from the medical institution providing this service.
If there are problems using the service, the medical institution will help solve them.
I could receive education in using this service.
Behavioral intention to use	20
If I get the chance, I intend to use the telemonitoring service.
I am willing to recommend the service to my friends or colleagues.
I am willing to use the measured results.
I want to get healthcare from the telemonitoring service.
Satisfaction	23,24
Overall, I am satisfied with this service.
I am satisfied with receiving blood glucose management services.
I am satisfied with periodically measuring my blood pressure or blood sugar level through this service.
I am satisfied with monitoring my blood glucose level.

Performance expectancy is defined as the degree to which an individual believes that using the system will help him or her improve his or her job performance.^19,20 Performance expectancy is equivalent to the construct of perceived usefulness in the TAM.²¹ In this study, performance expectancy was defined as the degree of improvement in glucose management associated with the use of a telemonitoring service. Second, effort expectancy is defined as the degree of ease associated with the use of the system.²⁰ In this study, effort expectancy was defined as the degree of ease that a patient associated with the use of the telemonitoring service. Third, social influence represents the individual's perception of the extent to which important others believe he or she should use the new system.²⁰ In this study, social influence was defined as the degree of important others' belief that a patient should use the telemonitoring service. Finally, facilitating conditions refer to the degree to which an individual believes that an organizational and technical infrastructure exists to support the use of the system.²⁰ Facilitating conditions can be interpreted in many ways depending on the research topic. In this study, facilitating conditions were defined as the degree to which a patient believed that there was an organizational and technical infrastructure supporting the use of telemonitoring services as well as resources offering the knowledge necessary to use these services.^20,22 In addition, the behavioral intention to use construct indicated the degree to which a patient intended to use the telemonitoring service.²⁰ Satisfaction described the degree of patients' satisfaction with the telemonitoring service.^23,24

Subjects

Data were obtained from randomized controlled trials in South Korea and China. This study only focused on the intervention group of each randomized controlled trial.

In South Korea, we conducted a survey and collected glucose data from patients with diabetes participating in a telemonitoring project that began in September 2012.²⁵ The project was a randomized controlled trial that aimed to demonstrate the importance of remote diabetes management in reducing the progression of diabetes complications in patients with type 2 diabetes.

In China, we conducted a survey and collected glucose data from patients with diabetes through a project implementing an Internet-based glucose monitoring system that was initiated in October 2013. This project used a randomized, open-label parallel-group design to evaluate the effects of telemonitoring.¹¹ The subjects were outpatients at the A Hospital at B University, China, who had been diagnosed with diabetes at least 1 year before the project.

To be included in the study, subjects had to have glycated hemoglobin (HbA1c) levels of 7.0–10.0%, have Internet connection at home, and provide voluntary informed consent. Subjects who were pregnant or who had a serious medical disease, serious diabetes complications, or aspartate transaminase or alanine transaminase levels ≥ 2.5 times the normal upper limit were excluded from the study. Subjects who had not taken their medication regularly in the 3 months before enrollment, those who were engaged in other clinical studies, and those without access to a computer were also excluded.

Telemonitoring service setting

The two telemonitoring projects had very similar study designs, as depicted in Figure 1.¹¹

Fig. 1.

Service flow.¹¹

In South Korea, patients were randomly allocated to the control or intervention group. Patients in the intervention group received the gateway device for free. Both groups were able to measure their blood glucose and blood pressure levels using mobile devices; however, only the intervention group could use the device to submit their data and to receive professional comments electronically. The primary outcome measure was the change in HbA1c levels, which were measured every 3 months. Patients were able to measure their blood glucose levels using the OneTouch UltraEasy glucose monitoring device (Johnson & Johnson Medical, Seoul, Korea) and their blood pressure using the OMRON HEM-7081-IT blood pressure gauge (OMRON, Seoul, Korea). Patients submitted their data using a home health gateway such as the HiCare 10 inch (Insung Information, Seoul, Korea). Patients could automatically submit their premeasured blood sugar and blood pressure levels and their dietary and exercise conditions to the Web site when they were logged into the gateway and attached to the glucose meter. Patients in the intervention group were required to submit data at least once per week. Patients were able to send questions to the medical team and to receive online consultations if needed. Once the data were transmitted, episodes of hyperglycemia and hypoglycemia and changes in BMI could be displayed. The medical team could use this information to provide customized health information to the patients through messages with recommendations or through online counseling. After patients submitted their data, the medical team could review their blood sugar levels, hypoglycemic episodes, and lifestyle factors. After reviewing the results, the medical team provided online counseling to provide blood pressure, blood glucose, nutrition, and exercise information. In addition, if patients had not recorded their blood sugar or blood pressure levels for 7 days or if they had not reported the causes of hypoglycemic and hyperglycemic episodes or of changes in BMI and HbA1c levels, they were classified as patients in need of intensive care. The medical team then provided customized interventions such as encouragement to use the telemonitoring service and exercise.

In China, patients with diabetes were randomly assigned to the control or intervention group. The intervention group received blood sugar monitoring through the Internet and visited the hospital every 3 months for laboratory testing and a clinical examination to evaluate the safety and efficacy of their treatment. Subjects in the intervention group were also provided glucometers (Myglucohealth OneTouch UltraEasy, Life Scan, US) and 150 testing strips for free. Patients in the intervention group received specific training in the system after registering at the project Web site. The subjects then measured their blood sugar levels using a system that was connected to their computers, and the results were automatically sent and saved to an online server. Subjects were able to view changes in their blood sugar levels as well as receive messages from the medical team through their computers. The intervention subjects uploaded their blood sugar data according to an established schedule, and the medical staff checked their health conditions on a regular basis. For the first 3 months, the nurses provided recommendations for blood sugar control on a weekly basis; biweekly recommendations were then provided for the remaining 3 months. If subjects failed to upload their blood sugar data for more than 1 week, they were sent an individualized warning message. Patients in the intervention group were required to submit their blood sugar data at least twice per week, and the number of submissions was adjusted according to the severity of each subject's condition. Subjects who completed less than half of the recommended tests or who exhibited low blood sugar levels were provided consultations through text messages or phone calls.

The South Korean medical team provided the clinical research protocol and educated the Chinese medical team. They also determined the clinical research guidelines, including the subjects' blood sugar level goals, the number of blood sugar level tests, and the medical feedback given to the subjects based on their results. They served as the system's control center, technical support, and health coordinating center.

Research design

The study in South Korea was conducted over a 12-month period. Patients visited the hospital every 3 months after completing their first visit for screening and their second visit 1 month later. Patient surveys were collected at each patient's last visit in 2013. A total of 81 valid responses were collected. HbA1c values were obtained from 81 registered patients with diabetes to assess their improvement in HbA1c levels. The project was conducted with approval from the Institutional Review Board of the D University of Korea. The study in China was conducted over a 6-month period. Patients visited the hospital every 3 months. Patient surveys were collected at each patient's last visit in 2014. A total of 92 valid responses were obtained. HbA1c values were collected from 92 registered patients with diabetes to assess their improvement in HbA1c levels. The study was approved by the Chinese Research Ethics Committee, and informed consent was obtained from each patient. All responses were measured using a 7-point Likert scale (Table 2).

Table 2.

Service Follow-Up

PERIODS	SCREENING	FOLLOW-UP
VISIT	VISIT 1	VISIT 2	VISIT 3	VISIT 4	VISIT 5	VISIT 6
South Korea	1 month before	First day	3 months	6 months	9 months	12 months
China	1 month before	First day	3 months	6 months	—	—

Statistical analyses

We used chi-square tests to compare the general characteristics by group. This study used two independent sample t-tests to compare the patients' perceptions of telemonitoring for glucose control. Multiple regression analysis was used to determine the factors that had an impact on behavioral intentions to use telemonitoring, and Wilcoxon signed rank test was used to assess the differences in patients' HbA1c levels from baseline to follow-up (at 3 months and 6 months). SPSS, version 21.0, software (SPSS, Inc., Chicago, IL) was used to perform the statistical analyses in this study.

Results

Sample characteristics

In South Korea, 51 respondents (63%) were male, 64 (79%) were over 50 years old, and 33 (40.7%) had an average monthly salary of over $4,000 (approximately $1,826 in USD at the exchange rate in 2014). In China, 48 respondents (52.2%) were male, 66 (71.7%) were over 50 years old, and 40 (43.5%) had an average monthly salary of over $4,000. Gender, age, and average monthly salary differed significantly between the groups.

In addition, there was a significant difference in the type of job (p < 0.001), compliance to telemonitoring (p < 0.001), and the amount participants were willing to pay for the telemonitoring service (p < 0.01). In South Korea, 27 participants (33.3%) were self-employed and 37 (45.7%) used the telemedicine service device on a daily basis. Additionally, 49 respondents (60.5%) thought that $4.5–$18 was an appropriate price for a monthly payment for the monitoring service. In contrast, 40 participants (43.5%) were office workers in China, and 72 (78.3%) used the telemedicine service device 1–2 times a week. Thirty-five respondents (38.0%) thought that $4.5–$9 was an appropriate price for monthly payments for the monitoring service (Table 3).

Table 3.

Respondent Characteristics

VARIABLE, CHARACTERISTIC	SOUTH KOREA	CHINA	p
VARIABLE, CHARACTERISTIC	FREQUENCY (%)	FREQUENCY (%)	p
Gender			0.271
Male	51(63.0)	48(52.2)
Female	30(37.0)	44(47.8)
Age (years)			0.248
Below 50	17(21.0)	26(28.3)
50–59	32(39.5)	36(39.1)
Over 60	32(39.5)	30(32.6)
Average monthly salary			0.248
Under 2,000	24(29.6)	26(28.3)
$2,000–$3,000	24(29.6)	26(28.3)
Over $4,000	33(40.7)	40(43.5)
Job			0.000^**
Office worker	16(19.8)	40(43.5)
Self-employed	27(33.3)	8(8.7)
Housewife	16(19.8)	1(1.1)
Other	22(27.2)	43(46.7)
Compliance			0.000^**
Every day	37(45.7)	1(1.1)
5–6 times a week	18(22.2)	5(5.4)
3–4 times a week	16(19.8)	14(15.2)
1–2 times a week	10(12.3)	72(78.3)
Amount willing to pay for service			0.004^*
Under $4.5	12(14.8)	29(31.5)
$4.5–$9	24(29.6)	35(38.0)
$9–$18	25(30.9)	20(21.7)
$18–$27	10(12.3)	3(3.3)
Over $28	10(12.3)	5(5.4)
Total	81(100)	92(100)

The exchange rate for Korean won to U.S. dollars was 1,826 wons in 2014.

t0.01 = 2.576, ^**t0.001 = 3.291.

Reliability and validity

To test the construct validity, principal component analysis with varimax rotation for each construct was performed. As shown in Table 4, the cross-loadings were lower than the corresponding factor loadings. Six factors emerged with no cross-construct loadings above 0.50. The pattern of loadings and cross-loadings supported the discriminant validity and internal consistency. The instrument also demonstrated convergent validity, with factor loadings exceeding 0.50 for each construct (see bold values in Table 4). The results identified twelve factors with eigenvalues greater than 1.0 that accounted for 91.567% of the total variance. In addition, the communality ranged from 0.639 to 0.930, with all items achieving the 0.50 threshold. Therefore, these results confirmed that the six constructs were distinct unidimensional scales.

Table 4.

Loadings, Cross-Loadings, and Reliability

VARIABLE	1	2	3	4	5	6	COMMUNALITY	CRONBACH'S A
Sat01	0.800	0.232	0.070	0.264	0.272	0.194	0.879	0.924
Sat02	0.739	0.202	0.241	0.307	0.271	0.180	0.846
Sat04	0.723	0.233	0.387	0.137	0.104	0.300	0.839
Sat03	0.675	0.218	0.236	0.264	0.074	0.452	0.846
BI01	0.297	0.768	0.342	0.247	0.212	0.210	0.945	0.972
BI02	0.296	0.708	0.322	0.295	0.290	0.268	0.935
BI03	0.272	0.701	0.353	0.196	0.363	0.255	0.925
BI04	0.235	0.653	0.393	0.224	0.314	0.368	0.921
SI01	0.212	0.316	0.795	0.232	0.286	0.183	0.953	0.973
SI02	0.255	0.351	0.770	0.229	0.303	0.164	0.946
SI03	0.302	0.421	0.657	0.240	0.357	0.226	0.936
EE01	0.286	0.208	0.257	0.821	0.215	0.200	0.952	0.928
EE02	0.257	0.196	0.278	0.802	0.171	0.282	0.935
EE03	0.394	0.340	0.056	0.618	0.373	0.195	0.833
FC01	0.217	0.351	0.339	0.255	0.719	0.255	0.933	0.971
FC02	0.257	0.304	0.395	0.278	0.710	0.239	0.953
FC03	0.288	0.330	0.413	0.276	0.670	0.219	0.936
PE01	0.358	0.268	0.214	0.276	0.243	0.751	0.945	0.960
PE02	0.363	0.305	0.163	0.254	0.280	0.749	0.956
PE03	0.528	0.286	0.258	0.318	0.186	0.580	0.899
Eigenvalue	3.709	3.377	3.150	2.814	2.682	2.581
% of Variance	18.545	16.883	15.750	14.071	13.411	12.907
Cumulative%	18.545	35.427	51.178	65.249	78.660	91.567

Extraction Method: Principal Component Analysis.

Rotation Method: Varimax with Kaiser Normalization.

Values shown in bold signify the loading value is greater than 0.5 and higher than loading value for other factors.

Sat, satisfaction; BI, behavioral intention to use telemonitoring; SI, social influence; EE, effort expectancy; FC, facilitating conditions; PE, performance expectancy.

The internal consistency reliability was evaluated using Cronbach's alpha. As shown in Table 4, the internal consistency values for all constructs were significant and ranged from 0.924 to 0.972 (0.924 for satisfaction, 0.972 for behavioral intention to use telemonitoring, 0.973 for social influence, 0.928 for effort expectancy, 0.971 for facilitating conditions, and 0.960 for performance expectancy). The internal consistency values for all the constructs were greater than 0.70.²⁶ Therefore, Cronbach's alpha values indicated that all the constructs were reliable.

Comparison of patients' perceptions between South Korea and China

The mean score of the six variables was higher in China than in South Korea. There was a significant difference between Chinese and South Korean participants in satisfaction (t = −7.138, p < 0.001), behavioral intention to use telemonitoring (t = −5.739, p < 0.001), social influence (t = −5.937, p < 0.001), facilitating conditions (t = −6.626, p < 0.001), performance expectancy (t = −6.472, p < 0.001), and effort expectancy (t = −7.696, p < 0.001). Despite these significant differences, both South Korean and Chinese patients had a high positive perception of telemonitoring (Table 5).

Table 5.

Perceptions of Telemonitoring in South Korea and China

	SOUTH KOREA		CHINA		T-TEST
VARIABLES	MEAN	SD	MEAN	SD	T-VALUE	p-VALUE
Sat	5.96	1.25	6.91	0.25	−7.138	0.000^*
BI	5.68	1.58	6.74	0.76	−5.739	0.000^*
SI	5.72	1.24	6.62	0.72	−5.937	0.000^*
FC	5.72	1.45	6.85	0.71	−6.626	0.000^*
EE	6.15	1.20	6.97	0.18	−6.472	0.000^*
PE	5.65	1.44	6.86	0.39	−7.696	0.000^*

t0.001 = 3.291.

SD, standard deviation.

Factors influencing behavioral intention to use telemonitoring

Multiple regression analysis was performed with behavioral intention to use telemonitoring as the dependent variable and performance expectancy, effort expectancy, social influence, and facilitating conditions as independent variables. A p-value of less than 0.05 was considered to be statistically significant.

In South Korea, the results of the multiple regression analysis indicated that performance expectancy and social influence contributed significantly to behavioral intention to use telemonitoring (F = 70.676, p = 0.000). The coefficient of determination (R ² ) for this model was 0.788, indicating that 78% of the variation in user satisfaction could be explained by these four independent variables. Table 4 shows that performance expectancy (t = 4.550, p = 0.001) and social influence (t = 3.587, p = 0.001) were significantly associated with behavioral intention to use telemonitoring.

By contrast, the results of the multiple regression analysis in China indicated that effort expectancy and facilitating conditions contributed significantly to behavioral intention to use telemonitoring (F = 107.321, p = 0.000). The coefficient of determination (R ²) for this model was 0.831, indicating that 83% of the variation in user satisfaction could be explained by these four independent variables. Effort expectancy (t = 1.950, p = 0.05) and facilitating conditions (t = 7.304, p = 0.001) were significantly associated with behavioral intention to use telemonitoring (Table 6).

Table 6.

Factors Influencing Behavioral Intention to Use Telemonitoring

	NONSTANDARDIZED COEFFICIENTS		STANDARDIZED COEFFICIENTS
INDEPENDENT VARIABLES		B	STANDARD ERROR	β	t-VALUE	p-VALUE
Total (n = 173)	(Constant)	−1.422	0.337	—	−4.224	0.000
	PE	0.384	0.082	0.268	4.695	0.000^**
	EE	0.065	0.063	0.059	1.033	0.303
	SI	0.484	0.078	0.400	6.229	0.000^**
	FC	0.274	0.074	0.258	3.677	0.000^**
South Korea(n = 81)	(Constant)	−1.524	0.464	—	−3.286	0.002
	PE	0.555	0.122	0.420	4.550	0.000^**
	EE	0.094	0.094	0.086	1.005	0.318
	SI	0.430	0.120	0.336	3.587	0.001^**
	FC	0.140	0.107	0.128	1.317	0.192
	By multiple regression analysis: R ² (adjusted R ²) = 0.788 (0.777)
China (n = 92)	(Constant)	−0.092	1.397	−	−0.066	0.948
	PE	−0.122	0.187	−0.029	−0.654	0.515
	EE	0.179	0.092	0.093	1.950	0.054^*
	SI	0.130	0.106	0.125	1.233	0.221
	FC	0.817	0.112	0.766	7.304	0.000^**
	By multiple regression analysis: R ² (adjusted R ²) = 0.831 (0.824)

t0.05 = 1.960, ^**t0.001 = 3.291.

Values shown in bold signify significant factors with intention to use telemonitoring.

Improvements in HbA1c levels

We assessed the differences in HbA1c levels from baseline to follow-up (at 3 and 6 months) in South Korean and Chinese patients with diabetes (Table 7). To examine the effects of age and telemonitoring on the subjects' blood sugar levels, this study divided each group into three subgroups according to age: below 40 years old, 50–59 years old, and over 60 years old.

Table 7.

HbA1c Improvements from Baseline to Follow-Up

	AGE	HBA1C (BASELINE)	HBA1C (3 MONTHS)	p-VALUE	HBA1C (6 MONTHS)	p-VALUE
South Korea (n = 81)	Below 50 years (n = 17)	7.86 ± 0.87	7.61 ± 0.72	0.133	7.51 ± 0.73	0.052 ^*
	50–59 years (n = 32)	7.64 ± 0.75	7.56 ± 0.81	0.503	7.34 ± 0.95	0.169
	Over 60 years (n = 32)	7.63 ± 0.55	7.52 ± 0.62	0.083	7.50 ± 0.78	0.236
China (n = 92)	Below 50 years (n = 26)	7.85 ± 1.57	6.70 ± 0.60	0.000 ^**	6.69 ± 0.64	0.000 ^**
	50–59 years (n = 36)	7.89 ± 0.86	6.86 ± 0.78	0.000 ^**	6.64 ± 0.70	0.000 ^**
	Over 60 years (n = 30)	7.80 ± 0.73	6.97 ± 0.65	0.000 ^**	6.86 ± 0.68	0.000 ^**

Data are presented as the mean ± SD or number (%) unless otherwise indicated.

t0.05 = 1.960, ^**t0.001 = 3.291.

HbA1c, Hemoglobin A1c.

In South Korea, patients who were younger than 50 years old exhibited a significant decrease in HbA1c levels at the 6-month follow-up (p = 0.053, p < 0.05). However, there were no significant differences in fasting HbA1c levels in the other groups. In China, subjects in all age groups showed a significant decrease in HbA1c levels at both follow-up points.

Discussion

This study was an international study on telemonitoring for glucose management conducted in South Korea and China. Based on our findings, we have reached the following conclusions.

Although there was a significant difference in the perceptions of telemonitoring between patients in China and South Korea, patients with diabetes in both countries had high levels of satisfaction, behavioral intentions to use telemonitoring, social influences, effort expectancies, facilitating conditions, and performance expectations for these services. The Korean patients with diabetes were more compliant with the service and willing to pay more for the service than the Chinese patients.

Different factors influenced Korean and Chinese patients' with diabetes behavioral intention to use telemonitoring. For Korean patients with diabetes, performance expectancy and social influence were associated with behavioral intention to use telemonitoring. In other words, Korean patients with diabetes believed that it was their first priority to improve their glucose control after using the telemonitoring service. Telemonitoring services must be supportive for glucose management and should be designed to support self-care and glucose monitoring. To accomplish this, diverse interventions and service protocols need to be developed with healthcare providers such as physicians, nurses, dieticians, and physical therapists. In addition, social influence had a strong influence on the behavioral intention to use telemonitoring among patients with diabetes in Korea.

In the UTAUT model, social influence refers to the individual's perception of whether important others believe that he or she should use the new system.²⁰ In general, important others as a source of social influence refer to organizations, supervisors, or coworkers. In telemonitoring services, the meaning differs depending on the user. To a patient with diabetes, important others may suggest their family members and partners.²⁷ Furthermore, it can refer to friends, peers, and patients with similar diseases,²² as well as medical providers who have direct influence over the patients in the healthcare setting, such as physicians and nurses.^22,28 Thus, diabetes patients in Korea are influenced by many important others, including physicians, family members, friends, and patients with similar diseases.

In Chinese patients with diabetes, in contrast, effort expectancy and facilitating conditions were the biggest factors contributing to their behavioral intention to use telemonitoring. In this study, effort expectancy meant the degree of ease that the patient associated with the use of the telemonitoring service. In other words, for Chinese patients with diabetes, it was important to be able to easily learn to use and then operate the telemonitoring service. Previous research has reported that effort expectancy has a critical influence on technology adoption.^12,29 In this setting, facilitating conditions were also important contributors to behavioral intentions to use telemonitoring. Facilitating conditions in this study referred to the degree to which a patient believed that an organizational and technical infrastructure existed to support the use of a telemonitoring service and that resources that provided the knowledge necessary to use the service were available.²² Facilitating conditions are especially crucial when an individual begins to use a new technology. Furthermore, Chinese patients with diabetes, many of whom are in older generations, may find it difficult to use new telemonitoring services and devices. Therefore, Chinese patients with diabetes need assistance and resources, such as knowledge, training, and appropriate equipment for telemonitoring services.

This study yielded very interesting results that indicated that different factors influence behavioral intentions to use telemonitoring among patients with diabetes in South Korea and China. Although South Korea and China already have an advanced telemonitoring service that uses information technology, they each have different environments that affect the acceptability of healthcare-related information technology. For these reasons, although the telemonitoring services were similar in form, the acceptance of telemonitoring services by patients with diabetes and their behaviors differed by country. Thus, telemonitoring services need to be designed to reflect patient groups and characteristics that are specific to each country. To develop telemonitoring services that account for regional differences and characteristics, there is a need to understand a variety of telemonitoring services.

Korean patients with diabetes who were below 50 years of age showed a significant decrease in HbA1c levels at the 6-month follow-up. Chinese patients of all ages experienced improvements in HbA1c levels at all follow-up points. These results showed that telemonitoring is an effective intervention in blood sugar management, which is consistent with previous research that has reported on the use of electronic devices for diabetes control.^3,5,6,8,30 Thus, telemonitoring services for blood sugar management should be actively developed and expanded among patients with diabetes.

This study had several limitations.

First, this study focused on South Korea and China. To determine the distinct characteristics of and differences between countries, future research should study a number of different countries. Second, we used only 173 patients in randomized controlled trial to control system characteristic because system characteristic could affect telemonitoring acceptance. Thus, they used very similar telemonitoring service. If telemonitoring will be deployed on a commercial scale, future research could take a larger sample. Third, we examined factors that influenced the acceptance of telemonitoring services, and the results for South Korea and China differed. These different influential factors could have resulted from individuals' intentions and behavioral changes. However, we did not examine these characteristics. According to the transtheoretical model, an individual's intentions and behavioral changes can be considered to occur in six stages: precontemplation, contemplation, preparation, action, maintenance, and occasionally relapse.³¹ This model demonstrates how people modify problematic behaviors or acquire positive behaviors, which could affect the acceptance of telemonitoring services. Thus, future studies should approach this topic using the transtheoretical model. Finally, there are other clinical variables, such as total cholesterol, triglycerides (TG), high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol, which this study did not include as outcomes. Future studies should consider a variety of clinical variables.

Despite these limitations, this study yielded valuable information. We found that telemonitoring is an effective intervention for blood sugar management in patients with diabetes. In addition, we explored the distinct characteristics of and differences between two countries. Accordingly, this study provides practical guidance for developing telemonitoring services for glucose management by considering the distinct characteristics of a country.

Footnotes

Acknowledgments

This study was supported by the Technology Innovation Program (Number 10035059, Primary Care-Centered Smart Care Service pilot project) funded by the Ministry of Trade, Industry and Energy (South Korea).

Disclosure Statement

The authors declare no conflicts of interest.

References

Bell

, Fonda

, Walker

, Schmidt

, Vigersky

. Mobile phone-based video messages for diabetes self-care support. J Diabetes Sci Technol, 2012; 6:310–319.

Shetty

, Chamukuttan

, Nanditha

, Raj

, Ramachandran

. Reinforcement of adherence to prescription recommendations in Asian Indian diabetes patients using short message service (SMS)–a pilot study. J Assoc Phys India, 2011; 59:711–714.

Curran

, Nichols

, Xie

, Harper

. An intensive insulinotherapy mobile phone application built on artificial intelligence techniques. J Diabetes Sci Technol, 2010; 4:209–220.

Hanley

, Fairbrother

, McCloughan

, Pagliari

, Paterson

, Pinnock

, et al. Qualitative study of telemonitoring of blood glucose and blood pressure in type 2 diabetes. BMJ Open, 2015; 5:e008896.

Levin

, Madsen

, Petersen

, Wanscher

, Hangaard

. Telemedicine diabetes consultations are cost-effective, and effects on essential diabetes treatment parameters are similar to conventional treatment: 7-year results from the Svendborg Telemedicine Diabetes Project. J Diabetes Sci Technol, 2013; 7:587–595.

Kulzer

, Daenschel

, Siegel

, Schramm

, Parkin

, et al. Integrated Personalized Diabetes Management (PDM) Design of the ProValue Studies: Prospective, Cluster-Randomized, Controlled, Intervention Trials for Evaluation of the Effectiveness and Benefit of PDM in Patients With Insulin-Treated Type 2 Diabetes. J Diabetes Sci Technol, 2016; 10:772–781.

Brown-Connolly

, Concha

, English

. Mobile health is worth it! Economic benefit and impact on health of a population-based mobile screening program in New Mexico. Telemed e-Health, 2014; 20:18–23.

Leng

, Jundong

, Wei

, Pin

, Ming

, Chen

. Telehealth for improved glycaemic control in patients with poorly controlled diabetes after acute hospitalization–a preliminary study in Singapore. J Telemed Telecare, 2014; 20:317–323.

Schellini

, Carvalho

GMd

, Rendeiro

, Padovani

, Hirai

. Prevalence of diabetes and diabetic retinopathy in a Brazilian population. Ophthalmic Epidemiol, 2014; 21:33–38.

10.

Rho

, Kim

H-S

, Cho

J-H

, Yoon

K-H

, Mun

, et al. Exploring the relationship among user satisfaction, compliance, and clinical outcomes of telemedicine services for glucose control. Telemed e-Health, 2014; 20:712–720.

11.

Kim

H-S

, Sun

, Yang

, Sun

, Li

, Choi

, et al. Randomized, open-label, parallel group study to evaluate the effect of internet-based glucose management system on subjects with diabetes in China. Telemed e-Health, 2016; 22:666–674.

12.

Rho

, Kim

, Chung

, Choi

. Factors influencing the acceptance of telemedicine for diabetes management. Cluster Comput, 2015; 18:321–331.

13.

Hennington

, Janz

. Information systems and healthcare XVI: Physician adoption of electronic medical records: Applying the UTAUT model in a healthcare context. Commun Assoc Inform Syst, 2007; 19:5.

14.

Schaper

, Pervan

. ICT and OTs: A model of information and communication technology acceptance and utilisation by occupational therapists. Int J Med Inform, 2007; 76:S212–S221.

15.

Arning

, Ziefle

. Different perspectives on technology acceptance: The role of technology type and age. Symposium of the Austrian HCI and Usability Engineering Group: Berlin Heidelberg, Springer, 2009; 20–41.

16.

Chang

I-C

, Hwang

H-G

, Hung

W-F

, Li

Y-C

. Physicians' acceptance of pharmacokinetics-based clinical decision support systems. Expert Syst Appl, 2007; 33:296–303.

17.

Nuq

. Innovation adoption (behavioral intention) for eHealth services in developing countries: A conceptual framework. J Int Univ Genava Bus Rev, 2009; 2:75–80.

18.

Venkatesh

, Sykes

, Zhang

. 'Just what the doctor ordered': A revised UTAUT for EMR system adoption and use by doctors. System Sciences (HICSS), 2011 44th Hawaii International Conference on: IEEE, New York, 2011; 1–10.

19.

van Bon

, Kohinor

, Hoekstra

, von Basum

, DeVries

. Patients' perception and future acceptance of an artificial pancreas. J Diabetes Sci Technol, 2010; 4:596–602.

20.

Venkatesh

, Morris

, Davis

. User acceptance of information technology: Toward a unified view. MIS Q, 2003:425–478.

21.

Kim

, Mirusmonov

, Lee

. An empirical examination of factors influencing the intention to use mobile payment. Comput Hum Behav, 2010; 26:310–322.

22.

Basoglu

, Daim

, Topacan

. Determining patient preferences for remote monitoring. J Med Syst, 2012; 36:1389–1401.

23.

Alaloola

, Albedaiwi

. Patient satisfaction in a Riyadh tertiary care centre. Int J Health Care Qual Assur, 2008; 21:630–637.

24.

Cardozo

, Steinberg

. Telemedicine for recently discharged older patients. Telemed e-Health, 2010; 16:49–55.

25.

Rho

, Kim

H-S

, Yoon

K-H

, Choi

. Compliance Patterns and Utilization of e-Health for Glucose Monitoring: Standalone Internet Gateway and Tablet Device. Telemed e-Health. 2017; 23:298–304.

26.

Nunnally

, Bernstein

, Berge JMt. Psychometric theory. JSTOR, New York: McGraw-Hill, 1967.

27.

Heart

, Kalderon

. Older adults: Are they ready to adopt health-related ICT?. Int J Med Inform, 2013; 82:e209–e231.

28.

Cranen

, Drossaert

, Brinkman

, Braakman-Jansen

, Ijzerman

, Vollenbroek-Hutten

. An exploration of chronic pain patients' perceptions of home telerehabilitation services. Health Expect, 2012; 15:339–350.

29.

Plouffe

, Hulland

, Vandenbosch

. Research report: Richness versus parsimony in modeling technology adoption decisions—understanding merchant adoption of a smart card-based payment system. Inform Syst Res, 2001; 12:208–222.

30.

Istepanian

, Zitouni

, Harry

, Moutosammy

, Sungoor

, Tang

, et al. Evaluation of a mobile phone telemonitoring system for glycaemic control in patients with diabetes. J Telemed Telecare, 2009; 15:125–128.

31.

Corcoran

Theories and models in communicating health messages. Communicating health: Strategies for health promotion. London, SAGE. 2007; 5–31.