Develop a PowerPoint Presentation in regards to the issue of Advanced Practice: Telehealth in Managing Chronic Diseases: Diabetes and Hypertension.

Citation: Calderón-Anyosa, R.;

Tincopa, J.P.; Raza, M.; Cárcamo, C.P.

Randomized Controlled Trial of

Home Telemonitoring of Blood

Pressure with an Adapted

Tensiometer with SMS Capability.

Eur. J. Investig. Health Psychol. Educ.

2023, 13, 440–449. https://doi.org/

10.3390/ejihpe13020033

Academic Editor: María del Mar

Molero Jurado

Received: 6 January 2023

Revised: 9 February 2023

Accepted: 10 February 2023

Published: 12 February 2023

Copyright: © 2023 by the authors.

Licensee MDPI, Basel, Switzerland.

This article is an open access article

distributed under the terms and

conditions of the Creative Commons

Attribution (CC BY) license (https://

creativecommons.org/licenses/by/

4.0/).

Article

Randomized Controlled Trial of Home Telemonitoring of Blood Pressure with an Adapted Tensiometer with SMS Capability Renzo Calderón-Anyosa 1, Jean Pierre Tincopa 1,2,* , Mabel Raza 3 and Cesar P. Cárcamo 1

1 Facultad de Salud Pública y Administración, Universidad Peruana Cayetano Heredia, Lima 15102, Peru 2 Digital Transformation Research Center, Universidad Norbert Wiener, Lima 15046, Peru 3 Facultad de Ciencias y Filosofía, Universidad Peruana Cayetano Heredia, Lima 15102, Peru * Correspondence: [email protected]

Abstract: Despite being a public health problem, less than a third of hypertensive patients manage to control blood pressure (BP). In this paper, we conducted a two-arm randomized controlled trial to investigate the efficacy of an SMS-based home BP telemonitoring system compared to usual care in patients with uncontrolled hypertension from a primary care center. This study was conducted between April and August 2018. Participants in the intervention arm used a custom-designed telemonitoring device for two weeks and were followed up for two additional weeks; controls were followed for 4 weeks. The main objective of this study is to evaluate the impact on blood pressure of a telemonitoring system using a blood pressure monitor adapted to send data via SMS to health providers in primary care centers for 4 weeks. In this trial, 38 patients were included in the analysis (18 in each arm), 68% were women, and the mean age was 68.1 [SD: 10.8 years], with no differences between arms. Among the results we found was that There was no significant difference in the change in systolic BP values between the control and intervention arm (−7.2 [14.9] mmHg vs. −16.3 [16.7] mmHg; p = 0.09). However, we found a significant difference in the change of diastolic BP (−1.2 [6.4] mmHg vs. −7.2 [9.8] mmHg; for the control and intervention arms, respectively p = 0.03). With all this, we conclude that an SMS-based home BP telemonitoring system is effective in reducing diastolic BP by working in conjunction with primary care centers. Our findings represent one of the first interventions of this type in our environment, being an important alternative for the control of high blood pressure.

Keywords: telemedicine; hypertension; primary care; short message system; monitoring

1. Introduction

High blood pressure is a public health problem, resulting in 15 million deaths per year worldwide [1]. The prevalence in Latin America varies from 30 to 50%; only 23% of men and 35% of women with diagnosed hypertension have their blood pressure under control [2,3].

In Peru, it is estimated that only 39.3% of those with hypertension are under treatment, and only 20% have their blood pressure under control [4], but in rural areas, this value can be as low as 4.9% [5]. Due to the low level of control of hypertensive patients, several studies have proposed telemonitoring measures of blood pressure at home, which has managed to increase the proportion of controlled patients. Current hypertension clinical guidelines suggest the use of telemonitoring both for the diagnosis and for the treatment of high blood pressure [6,7].

Home blood pressure telemonitoring is defined as the process by which blood pressure readings at home are transmitted to a central health information center or electronic medical record for use by healthcare providers and patients [8]. There is a large variety of telemonitoring systems, with differences in measurement methods and communication systems, including data transmission via Bluetooth, Wi-Fi, and telephone lines, among

Eur. J. Investig. Health Psychol. Educ. 2023, 13, 440–449. https://doi.org/10.3390/ejihpe13020033 https://www.mdpi.com/journal/ejihpe

Eur. J. Investig. Health Psychol. Educ. 2023, 13 441

others. The use of these devices may be limited in some contexts when they rely on external devices such as smartphones with Bluetooth technology or Wi-Fi hot spots at home [9], emphasizing the need for simple, cost-effective systems that are easy to use and acceptable to both patients and providers.

An alternative for the implementation of telemonitoring at home is the adaptation of blood pressure monitors for home use, enabling them to send data through text messages (SMS), a technology with wide penetration [10]. Some studies have shown the feasibility of adapting blood pressure monitors to send blood pressure data via SMS [11,12]. Despite the technological advantages of using this telecommunication system, there is little information regarding its clinical impact. The main objective of this study is to evaluate the impact on blood pressure of a telemonitoring system using a blood pressure monitor adapted to send data via SMS to health providers in primary care centers for 4 weeks.

2. Materials and Methods 2.1. Study Design

A randomized controlled trial was performed. The intervention arm received the telemonitoring system, while the control arm continued with the usual management. The main outcome was the difference in systolic blood pressure (SBP) and diastolic blood pressure (DBP) at follow-up. The blood pressure monitors used in this study were adapted to send SMS using the services of a local mobile telecommunications provider.

The data were obtained confidentially. Personal identifiers were stored separately in a password-protected database to which only the researchers had access. The study protocol was registered on clinicaltrials.gov (NCT03524456).

2.2. Participants’ Enrollment

This study included patients with uncontrolled high blood pressure treated at the Condevilla health center in the district of San Martín de Porres, in the province of Lima. Patients with uncontrolled treated hypertension were defined as those with SBP above 130 mmHg or DBP above 80 mmHg [8].

The inclusion criteria were the following: (i) Patients older than 18 years; (ii) Diag- nosis of high blood pressure for at least 3 months; (iii) Uncontrolled blood pressure; and (iv) Under treatment with antihypertensive medication. Exclusion criteria were (i) Patients on hemodialysis or peritoneal dialysis for chronic kidney disease; (ii) Pregnant women, and (iii) Patients planning to travel or change of address within a month of enrollment.

2.3. Randomized Grouping

Patients were referred to the study by the Condevilla health center. If they met the inclusion and exclusion criteria, they were invited to participate in the study and asked to provide informed consent. Randomization was performed in complete blocks of size 4.

2.4. Sample Size

For a significance level of 95% and 80% power, a standard deviation of 10 and 12 mmHg for the control and intervention arm and a sample size of 20 participants per arm would be required to detect differences of at least 10 mmHg [13].

2.5. Intervention 2.5.1. Development of the Telemonitoring System

A commercial Omron Series 10® blood pressure monitor was used, which has USB (universal serial bus) connectivity from the factory. The USB port built into the blood pressure monitor was used to link it to the data capture and delivery module (Figure 1). The monitor itself was not modified. The cost of this equipment was around 70 USD. The accuracy of this monitor is recognized by the AHA (American Heart Association), and it has been used in other studies [14]. The equipment is easy to use and was designed

Eur. J. Investig. Health Psychol. Educ. 2023, 13 442

for home use, with an easy-to-set cuff and a large number display. It shows the values of systolic and diastolic blood pressure, as well as heart rate, on an LCD screen.

Eur. J. Investig. Health Psychol. Educ. 2023, 13  442   

2.5. Intervention 

2.5.1. Development of the Telemonitoring System 

A commercial Omron Series 10® blood pressure monitor was used, which has USB 

(universal serial bus) connectivity from the factory. The USB port built into the blood pres‐

sure monitor was used to link it to the data capture and delivery module (Figure 1). The 

monitor itself was not modified. The cost of this equipment was around 70 USD. The ac‐

curacy of this monitor is recognized by the AHA (American Heart Association), and it has 

been used in other studies [14]. The equipment is easy to use and was designed for home 

use, with an easy‐to‐set cuff and a large number display. It shows the values of systolic 

and diastolic blood pressure, as well as heart rate, on an LCD screen. 

Figure 1. Total assembly of the telemonitoring equipment. (A) Connections between components 

(B) Final version with MDF casings. 

2.5.2. Hardware Development 

The data capture and delivery module used Arduino development boards. Arduino 

is an open‐source hardware and software company that designs and manufactures single‐

board microcontrollers and microcontroller kits for building digital devices and interac‐

tive objects. Arduino boards are commercially available, with a wide catalog of devices 

and kits [15]. 

Figure 1. Total assembly of the telemonitoring equipment. (A) Connections between components (B) Final version with MDF casings.

2.5.2. Hardware Development

The data capture and delivery module used Arduino development boards. Arduino is an open-source hardware and software company that designs and manufactures single- board microcontrollers and microcontroller kits for building digital devices and interactive objects. Arduino boards are commercially available, with a wide catalog of devices and kits [15].

2.5.3. Software Development

Arduino has its integrated development environment (Arduino Integrated Develop- ment Environment IDE. It is a cross-platform application (for Windows, macOS, and Linux). This platform allows users to develop, compile and upload programs to an Arduino board.

2.5.4. Telemonitoring Protocol

Participants in the intervention arm received the telemonitoring equipment and were asked to take at least two blood pressure measurements in the mornings and two in the evenings for two weeks. The blood pressure measurements were sent to the health person- nel at the Condevilla health center, who made clinical decisions based on the measurements. These decisions could be adjustments or changes in medication or calls to the patient for appointments at the center.

Participants in the control arm received the status quo, consisting of monitoring based on visits to the health center for blood pressure determinations and treatment indications.

Eur. J. Investig. Health Psychol. Educ. 2023, 13 443

2.5.5. Technical and Process Aspects

The technical and process aspects of home blood pressure monitoring are important to ensure accurate and reliable readings and to allow for effective monitoring and management of the patient’s blood pressure.

• Patient training: The patient is trained on how to properly use the blood pressure monitor and how to accurately record the readings.

• Equipment: A blood pressure monitor with telemonitoring capability, such as an adapted tensiometer with SMS capability, is provided to the patient.

• Monitoring frequency: The frequency of monitoring was determined and agreed upon between the patient and healthcare provider in two measurements in the morning and two measurements at night spaced out by 5 min each daily, after 1 min of rest.

• Data transmission: The patient takes their blood pressure readings, and the telemoni- toring capability of the blood pressure monitor transmits the data to their healthcare provider using via SMS.

• Review by a healthcare provider: The healthcare provider receives the transmitted data and reviews the readings to monitor the patient’s blood pressure and identify any potential issues.

• Follow-up: The healthcare provider may schedule a follow-up appointment with the patient, if necessary, based on the review of the transmitted data.

2.6. Study Setting

Outpatients of the health center were invited to participate in the study after verifying the inclusion and exclusion criteria. The patients were informed of the study objectives, procedures, and requirements. Consenting patients were asked to sign the informed consent form.

After randomization, those in the intervention arm were given a training session in the use of the telemonitoring system. They were provided with information on the proper placement of the cuff, as well as on the correct position of the body during blood pressure measurement. They were given an informative guide as well as a user manual.

Study personnel delivered the monitors to participants during home visits. The most suitable place for taking blood pressure was selected to suit the patient. After choosing the place, the signal of the equipment and the reception of the message were both verified through a blood pressure measurement (not considered in the analysis).

Participants skipping more than two consecutive blood pressure measurements were contacted by telephone to evaluate the reasons for the discontinuation and to encourage continuous monitoring. The decision to adjust the medication or call the patient was given by the health center’s physician. The final measurement was taken two weeks after completing the telemonitoring. For participants in the control arm, baseline and follow-up measurements were taken four weeks apart (Figure 2).

2.7. Statistical Analysis

The primary statistical analysis was performed using the Student’s t-test for two independent samples comparing the mean change in blood pressure between intervention and control arms. The change in blood pressure was the difference between the four-week value and the baseline value. Qualitative variables have been expressed as a percentage. A Fisher’s exact test, or a Chi2 test, was used for the comparison of proportions. For nor- mally distributed quantitative variables, means and ±standard deviations are presented. A comparison of means was performed using the Student’s t-test. Non-normally distributed quantitative variables were described with medians and interquartile ranges, and compar- isons were performed using the Mann–Whitney test. Associations with p values < 0.05 were considered statistically significant. Statistical analysis was performed using the STATA 15 (StataCorp, College Station, TX, USA).

Eur. J. Investig. Health Psychol. Educ. 2023, 13 444

Eur. J. Investig. Health Psychol. Educ. 2023, 13  444   

continuous monitoring. The decision to adjust the medication or call the patient was given 

by the health center’s physician. The final measurement was taken two weeks after com‐

pleting  the  telemonitoring. For participants  in  the control arm, baseline and  follow‐up 

measurements were taken four weeks apart (Figure 2). 

Figure 2. Design of the intervention. 

2.7. Statistical Analysis 

The primary statistical analysis was performed using the Student’s t‐test for two in‐

dependent samples comparing the mean change in blood pressure between intervention 

and control arms. The change in blood pressure was the difference between the four‐week 

value and the baseline value. Qualitative variables have been expressed as a percentage. 

A Fisher’s exact test, or a Chi2 test, was used for the comparison of proportions. For nor‐

mally distributed quantitative variables, means and ±standard deviations are presented. 

A comparison of means was performed using the Student’s t‐test. Non‐normally distrib‐

uted quantitative variables were described with medians and  interquartile ranges, and 

comparisons were performed using the Mann–Whitney test. Associations with p values < 

0.05 were considered statistically significant. Statistical analysis was performed using the 

STATA 15 (StataCorp, College Station, TX, USA). 

3. Results 

3.1. Recruitment of Participants 

The  recruitment  and  follow‐up of patients  took place between April  and August 

2018. A total of 223 patients were evaluated for eligibility, of which 183 patients were ex‐

cluded (134 did not present a diagnosis of arterial hypertension, 23 had less than 3 months 

of diagnosis, 20 were not on pharmacological at screening, and 6 patients did not agree to 

participate in the study). A total of 40 patients were randomized: 20 to the control arm 

and 20 to the intervention arm. One of the patients in the control arm withdrew before the 

second blood pressure measurement. Likewise, one patient in the intervention arm died 

days before the second blood pressure measurement was performed; thus, 38 participants 

were included in the final analysis, 19 in each arm (Figure 3). 

Figure 2. Design of the intervention.

3. Results 3.1. Recruitment of Participants

The recruitment and follow-up of patients took place between April and August 2018. A total of 223 patients were evaluated for eligibility, of which 183 patients were excluded (134 did not present a diagnosis of arterial hypertension, 23 had less than 3 months of diagnosis, 20 were not on pharmacological at screening, and 6 patients did not agree to participate in the study). A total of 40 patients were randomized: 20 to the control arm and 20 to the intervention arm. One of the patients in the control arm withdrew before the second blood pressure measurement. Likewise, one patient in the intervention arm died days before the second blood pressure measurement was performed; thus, 38 participants were included in the final analysis, 19 in each arm (Figure 3).

3.2. Baseline Characteristics 3.2.1. Demographics Data

The majority of participants (26; 68.4%) were female. The mean age of 68.1 ± 10.8 years, and the mean BMI was 29.5 ± 4.0 kg/m2. 14 (36.8%) of the participants were born in Lima), 22 (57.9%) were married or cohabiting, 24 (63.2%) had only primary education, and 20 (52.6%) were housewives. The only significant difference found within the demographic and clinical variables between the two study arms was marital status (p = 0.026).

3.2.2. Clinical Data

The mean time since diagnosis of high blood pressure was 11.1 ± 7.7 months, while the most frequent pharmacological treatment was angiotensin-converting enzyme inhibitors (ACEI) (n = 24; 63.2%). Likewise, it was found that the majority of participants had a family history of hypertension (n = 21; 55.3%). The most common comorbidity was Diabetes Mellitus (n = 9; 23.7%). No differences were found in clinical variables between the two study arms (p > 0.05) (Table 1).

Eur. J. Investig. Health Psychol. Educ. 2023, 13 445

Eur. J. Investig. Health Psychol. Educ. 2023, 13  445   

Figure 3. CONSORT flowchart of participating individuals throughout the essay. 

3.2. Baseline Characteristics 

3.2.1. Demographics Data 

The majority of participants  (26; 68.4%) were  female. The mean age of 68.1 ± 10.8 

years, and the mean BMI was 29.5 ± 4.0 kg/m2. 14 (36.8%) of the participants were born in 

Lima), 22 (57.9%) were married or cohabiting, 24 (63.2%) had only primary education, and 

20  (52.6%) were  housewives.  The  only  significant  difference  found within  the  demo‐

graphic and clinical variables between the two study arms was marital status (p = 0.026). 

3.2.2. Clinical Data 

The mean time since diagnosis of high blood pressure was 11.1 ± 7.7 months, while 

the most frequent pharmacological treatment was angiotensin‐converting enzyme inhib‐

itors (ACEI) (n = 24; 63.2%). Likewise, it was found that the majority of participants had a 

family history of hypertension (n = 21; 55.3%). The most common comorbidity was Diabe‐

tes Mellitus (n = 9; 23.7%). No differences were found in clinical variables between the two 

study arms (p > 0.05) (Table 1). 

Figure 3. CONSORT flowchart of participating individuals throughout the essay.

In the total sample, the mean SBP at the baseline measurement was 156.7 ± 13.7 mmHg, while the DBP was 86.6 ± 9.6 mmHg; there were no differences in blood pressure between control and intervention arms at baseline. Regarding the final measurement, the mean SBP was 144.9 ± 15.8 mmHg, and the DBP was 82.3 ± 10.6 mmHg; no differences between study arms were found (Table 2).

Eur. J. Investig. Health Psychol. Educ. 2023, 13 446

Table 1. General characteristics of the participants by study arms.

Control Intervention Total p

Female Sex 16 (84.2%) 10 (52.6%) 26 (68.4%) 0.079 Age (years) 68.7 ± 11.5 67.5 ± 10.3 68.1 ± 10.8 0.744

BMI (kg/m2) 29.4 ± 3.4 29.5 ± 4.6 29.5 ± 4.0 0.988 Born in Lima 5 (26.3%) 9 (47.4%) 14 (36.8%) 0.313 Marital status

Single 3 (15.8%) 2 (10.5%) 5 (13.2%) 0.026 Married/Cohabiting 7 (36.8%) 15 (79.0%) 22 (57.9%)

Divorced 1 (5.3%) 0 1 (2.6%) Widow/widower 8 (42.1%) 2 (10.5%) 10 (26.3%) Education level

Primary 13 (68.4%) 11 (57.9%) 24 (63.2%) 0.788 High school 5 (26.3%) 6 (31.6%) 11 (29.0%)

Superior 1 (3.3%) 2 (10.5%) 3 (7.9%) Occupation Housewife 11 (57.9%) 9 (47.4%) 20 (52.6%) 0.455

Casual work 2 (10.5%) 5 (26.3%) 7 (18.4%) Permanent job 6 (31.6%) 5 (26.3%) 11 (29.0%)

HTA diagnosis time (months) 10 (5–12) 10 (3–16) 10 (5–15) 0.918

Treatment ACEI 15 (79.0%) 9 (47.4%) 24 (63.2%) 0.248

ARA-II 3 (15.8%) 5 (26.3%) 8 (21.1%) CCB 0 1 (5.3%) 1 (2.6%)

ARA-II + CCB 1 (5.3%) 3 (15.8%) 4 (10.5%) ACEI + CCB + ARA-II 0 1 (5.3%) 1 (2.6%) Family history of high

blood pressure 10 (52.6%) 11 (57.9%) 21 (55.3%) 1

Comorbidities Diabetes 6 (31.6%) 3 (15.8%) 9 (23.7%) 0.447 Others 14 (73.7%) 12 (63.2%) 26 (68.4%) 0.728

HTA: Arterial hypertension, ACEI: Angiotensin-converting enzyme inhibitors, ARA-II: angiotensin II receptor antagonists, CCB: calcium channel blockers.

Table 2. Change in blood pressure values between the baseline measurement and the final measure- ment of the study arm.

Control Intervention Total p P MW

Baseline measurement (BM) SBP (mmHg) 157.0 ± 15.2 156.3 ± 12.4 156.7 ± 13.7 0.880 0.9301 DBP (mmHg) 84.8 ± 9.1 88.3 ± 9.9 86.6 ± 9.6 0.276 0.3128

Final measurement (FM) SPB (mmHg) 149.8 ± 17.4 140.1 ± 12.6 144.9 ± 15.8 0.056 0.111 DBP (mmHg) 83.6 ± 10.9 81.1 ± 10.3 82.3 ± 10.3 0.460 0.6180

FM–BM SBP (mmHg) −7.2 ± 14.9 −16.3 ± 16.7 −11.7 ± 16.3 0.087 0.0470 DBP (mmHg) −1.2 ± 6.4 −7.2 ± 9.8 −4.2 ± 8.7 0.032 0.0452

SBP: systolic blood pressure, DBP: diastolic blood pressure.

3.3. Primary Outcome

The difference in SBP values (final measurement—initial measurement) was −16.3 ± 16.7 mmHg in the intervention arm, while in the control arm, it was −7.2 ± 14.9 mmHg (p = 0.087). Similarly, the difference in DBP values was −7.2 ± 9.8 mmHg in the intervention arm, while in the control arm, it was −1.2 ± 6.4 mmHg (p = 0.032).

Eur. J. Investig. Health Psychol. Educ. 2023, 13 447

4. Discussion

Our study shows a reduction in SBP values and a significant reduction in DBP values in those participants who used the telemonitoring system compared to the control arm. These results show the potential benefit of including telemonitoring systems in primary care centers for the control of hypertensive patients, and if scaled up, these systems represent an important alternative strategy for public health.

This study shows that immediately after using the telemonitoring system, the reduc- tion in blood pressure values is greater than after discontinuing its use for two weeks. This could be because the effect is lost after the equipment use stops since the effect could be related to the measurement, visualization of results, and communication with the doctor.

The reduction in SBP and DBP values found in our study is consistent with that reported in previous studies. A meta-analysis that evaluated clinical trials involving telemonitoring systems for blood pressure control found an average reduction in SBP of −4.8 mmHg and DBP of −2.1 mmHg, with some studies showing reduction values as large as −25 mmHg and −15 mmHg for SBP and DBP, respectively [16]. Our results show an av- erage reduction in SBP of −16.3 ± 16.7 and DBP of −7.2 ± 9.8. This greater reduction could be because the baseline blood pressure values were higher in our population compared to most of the studies included in this meta-analysis.

A study conducted in the United States, which evaluated an intervention with a blood pressure telemonitoring system with medication adjustment by pharmacists, found after 6 months of follow-up, mean reductions of −6.0 (SBP) and −2.0 mmHg (DBP), significantly higher than those in controls [17]. Although the follow-up time was shorter in our study, suggesting that blood pressure goals can be achieved in a shorter follow-up time.

On the other hand, our results are comparable with studies that used more sophis- ticated telecommunication systems. A study conducted in Germany that evaluated a telemonitoring system using a blood pressure monitor connected to a mobile phone via Bluetooth found mean reductions in the intervention arm of −17.0 (SBP) and −9.0 mmHg DBP [18] significantly larger than those in the control arm. It should be noted that our sample consisted mainly of patients older than 70 years, while this study had a mean age of 55 years.

Age can represent a limiting factor for the use of some telecommunication systems, resulting in a barrier to the implementation of telemonitoring systems [8]. The simplifica- tion of processes and the elimination of the need for additional equipment represents an advantage of our study, allowing access to patients unfamiliar with modern technology.

Regarding the use of telemonitoring systems in primary care centers, information is scarce [19], and there are mainly studies that have evaluated their acceptability and feasibility [20,21] without presenting or evaluating their clinical effect. A study conducted in the United Kingdom, which evaluated a telemonitoring system at the first level of care, found reductions in SBP and DBP after 6 months of follow-up similar to those found in our study, showing that care at primary centers can be improved by the use of telemonitoring systems [22].

Regarding the causal relationship of the effectiveness of the telemonitoring system, we speculate that in our study, it could be due to an improvement in adherence to treatment, but we have the limitation of not having evaluated this parameter in this study.

5. Conclusions

Our study shows a reduction in SBP values and a significant reduction in DBP values in those participants who used the telemonitoring system compared to the control group. These results show the potential benefit of including telemonitoring systems in primary care centers for the control of hypertensive patients, and if scaled up, they represent an important alternative strategy for public health.

The telemonitoring system helps improve adherence to treatment, also improving disease awareness. When patients see the blood pressure values, they have an objec- tive parameter to base themselves on; likewise, knowing that there is someone on the

Eur. J. Investig. Health Psychol. Educ. 2023, 13 448

other side continuously reviewing the values helps them to continue and not stop taking the medication.

Part of the future work that could be generated from this study is to take it to rural environments where the only constant telecommunication signal is SMS, in addition to evaluating this tool for longer follow-up times and with a larger sample size.

Author Contributions: Conceptualization, R.C.-A. and M.R.; methodology, C.P.C.; software, J.P.T.; formal analysis, R.C.-A. and M.R.; writing—original draft preparation, J.P.T.; writing—review and editing, R.C.-A., M.R., C.P.C. and J.P.T. All authors have read and agreed to the published version of the manuscript.

Funding: This work was supported by CONCYTEC—FONDECYT (CONV-000229-2015).

Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Bioethics Committee of the Universidad Peruana Cayetano Heredia (SIDISI: 101180—Approved in November 2017).

Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments: This work was developed thanks to funding from the National Fund for Sci- entific, Technological and Technological Innovation Development (FONDECYT), an initiative of the National Council for Science, Technology and Technological Innovation (CONCYTEC) and the “Runachay” program—Training in the use of information and communication technologies for Global Health. The authors would like to express their gratitude to the personnel of the Condevilla Health Center for their support and collaboration during the study.

Conflicts of Interest: The authors declare no conflict of interest.

References 1. Beaney, T.; Schutte, A.E.; Tomaszewski, M.; Ariti, C.; Burrell, L.M.; Castillo, R.R.; Charchar, F.J.; Damasceno, A.; Kruger, R.;

Lackland, D.T.; et al. May Measurement Month 2017: An analysis of blood pressure screening results worldwide. Lancet Glob. Health 2018, 6, e736–e743. [CrossRef] [PubMed]

2. Lopez-Jaramillo, P.; Lopez-Lopez, J.; Cohen, D.; Alarcon-Ariza, N.; Mogollon-Zehr, M. Epidemiology of Hypertension and Diabetes Mellitus in Latin America. Curr. Hypertens. Rev. 2021, 17, 112–120. [CrossRef] [PubMed]

3. The Lancet Regional Health-Americas. Latin America and Caribbean’s path to improve hypertension control: Time for bolder, tougher actions. Lancet Reg. Health Am. 2022, 9, 100278. [CrossRef]

4. Villarreal-Zegarra, D.; Carrillo-Larco, R.M.; Bernabe-Ortiz, A. Short-term trends in the prevalence, awareness, treatment, and control of arterial hypertension in Peru. J. Hum. Hypertens. 2021, 35, 462–471. [CrossRef] [PubMed]

5. Elnaem, M.H.; Mosaad, M.; Abdelaziz, D.H.; Mansour, N.O.; Usman, A.; Elrggal, M.E.; Cheema, E. Disparities in Prevalence and Barriers to Hypertension Control: A Systematic Review. Int. J. Environ. Res. Public. Health 2022, 19, 14571. [CrossRef] [PubMed]

6. Pekmezaris, R.; Tortez, L.; Williams, M.; Patel, V.; Makaryus, A.; Zeltser, R.; Sinvani, L.; Wolf-Klein, G.; Lester, J.; Sison, C.; et al. Home Telemonitoring In Heart Failure: A Systematic Review And Meta-Analysis. Health Aff. (Millwood) 2018, 37, 1983–1989. [CrossRef]

7. Omboni, S.; McManus, R.J.; Bosworth, H.B.; Chappell, L.C.; Green, B.B.; Kario, K.; Logan, A.G.; Magid, D.J.; Mckinstry, B.; Margolis, K.L.; et al. Evidence and Recommendations on the Use of Telemedicine for the Management of Arterial Hypertension. Hypertension 2020, 76, 1368–1383. [CrossRef] [PubMed]

8. Nerenberg, K.A.; Zarnke, K.B.; Leung, A.A.; Dasgupta, K.; Butalia, S.; McBrien, K.; Harris, K.C.; Nakhla, M.; Cloutier, L.; Gelfer, M.; et al. Hypertension Canada’s 2018 Guidelines for Diagnosis, Risk Assessment, Prevention, and Treatment of Hypertension in Adults and Children. Can. J. Cardiol. 2018, 34, 506–525. [CrossRef]

9. Batista, E.; Moncusi, M.A.; López-Aguilar, P.; Martínez-Ballesté, A.; Solanas, A. Sensors for Context-Aware Smart Healthcare: A Security Perspective. Sensors 2021, 21, 6886. [CrossRef]

10. Ceballos, F.; Hernandez, M.A.; Olivet, F.; Paz, C. Assessing the use of cell phones to monitor health and nutrition interventions: Evidence from rural Guatemala. PLoS ONE 2020, 15, e0240526. [CrossRef]

11. Khoong, E.C.; Commodore-Mensah, Y.; Lyles, C.R.; Fontil, V. Use of Self-Measured Blood Pressure Monitoring to Improve Hypertension Equity. Curr. Hypertens. Rep. 2022, 24, 599–613. [CrossRef] [PubMed]

12. Allen, M.E.; Irizarry, T.; Einhorn, J.; Kamarck, T.W.; Suffoletto, B.P.; Burke, L.E.; Rollman, B.L.; Muldoon, M.F. SMS-Facilitated Home Blood Pressure Monitoring: A Qualitative Analysis of Resultant Health Behavior Change. Patient Educ. Couns. 2019, 102, 2246–2253. [CrossRef] [PubMed]

Eur. J. Investig. Health Psychol. Educ. 2023, 13 449

13. Gopalakrishnan, S.; Savitha, A.K.; Rama, R. Evaluation of inter-arm difference in blood pressure as predictor of vascular diseases among urban adults in Kancheepuram District of Tamil Nadu. J. Fam. Med. Prim. Care 2018, 7, 142–146. [CrossRef] [PubMed]

14. Pletcher, M.J.; Fontil, V.; Carton, T.; Shaw, K.M.; Smith, M.; Choi, S.; Todd, J.; Chamberlain, A.M.; O’Brien, E.C.; Faulkner, M.; et al. The PCORnet Blood Pressure Control Laboratory. Circ. Cardiovasc. Qual. Outcomes 2020, 13, e006115. [CrossRef] [PubMed]

15. Arduino—Home. Available online: https://www.arduino.cc/ (accessed on 5 September 2022). 16. Duan, Y.; Xie, Z.; Dong, F.; Wu, Z.; Lin, Z.; Sun, N.; Xu, J. Effectiveness of home blood pressure telemonitoring: A systematic

review and meta-analysis of randomised controlled studies. J. Hum. Hypertens. 2017, 31, 427–437. [CrossRef] [PubMed] 17. Li, X.; Hu, J.; Yao, Y.; Zuo, C.; Wang, Z.; Li, X.; Lv, Q. Evaluation of pharmacist-led telemedicine medication management for

hypertension established patients during COVID-19 pandemic: A pilot study. Front. Public Health 2022, 10. [CrossRef] 18. Park, S.-H.; Shin, J.-H.; Park, J.; Choi, W.-S. An Updated Meta-Analysis of Remote Blood Pressure Monitoring in Urban-Dwelling

Patients with Hypertension. Int. J. Environ. Res. Public. Health 2021, 18, 10583. [CrossRef] 19. Hanley, J.; Pinnock, H.; Paterson, M.; McKinstry, B. Implementing telemonitoring in primary care: Learning from a large

qualitative dataset gathered during a series of studies. BMC Fam. Pract. 2018, 19, 118. [CrossRef] 20. Baratta, J.; Brown-Johnson, C.; Safaeinili, N.; Rosas, L.G.; Palaniappan, L.; Winget, M.; Mahoney, M. Patient and Health

Professional Perceptions of Telemonitoring for Hypertension Management: Qualitative Study. JMIR Form. Res. 2022, 6, e32874. [CrossRef]

21. Pekmezaris, R.; Williams, M.S.; Pascarelli, B.; Finuf, K.D.; Harris, Y.T.; Myers, A.K.; Taylor, T.; Kline, M.; Patel, V.H.; Murray, L.M.; et al. Adapting a home telemonitoring intervention for underserved Hispanic/Latino patients with type 2 diabetes: An acceptability and feasibility study. BMC Med. Inform. Decis. Mak. 2020, 20, 324. [CrossRef]

22. Hammersley, V.; Parker, R.; Paterson, M.; Hanley, J.; Pinnock, H.; Padfield, P.; Stoddart, A.; Park, H.G.; Sheikh, A.; McKinstry, B. Telemonitoring at scale for hypertension in primary care: An implementation study. PLoS Med. 2020, 17, e1003124. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

RESEARCH ARTICLE

Telemedicine management of type 2 diabetes

mellitus in obese and overweight young and

middle-aged patients during COVID-19

outbreak: A single-center, prospective,

randomized control study

Wenwen Yin1,2☯, Yawen Liu3☯, Hao Hu2, Jin Sun2, Yuanyuan Liu2, Zhaoling WangID 2*

1 China University of Mining and Technology, Xuzhou, Jiangsu Province, China, 2 Department of

Endocrinology, The First People’s Hospital of Xuzhou, Xuzhou, Jiangsu Province, China, 3 Department of

Intensive Care Unit, The First People’s Hospital of Xuzhou, Xuzhou, Jiangsu Province, China

☯ These authors contributed equally to this work.

* [email protected]

Abstract

Objective

The coronavirus disease-2019 (COVID-19) pandemic severely affected the disease man-

agement of patients with chronic illnesses such as type 2 diabetes mellitus (T2DM). This

study aimed to assess the effect of telemedicine management of diabetes in obese and

overweight young and middle-aged patients with T2DM during the COVID-19 pandemic.

Methods

A single-center randomized control study was conducted in 120 obese or overweight (body

mass index [BMI]� 24 kg/m2) young and middle-aged patients (aged 18–55 years) with

T2DM. Patients were randomly assigned to the intervention (telemedicine) or control (con-

ventional outpatient clinic appointment) group. After baseline assessment, they were home

isolated for 21 days, received diet and exercise guidance, underwent glucose monitoring,

and followed up for 6 months. Glucose monitoring and Self-Rating Depression Scale (SDS)

scores were evaluated at 22 days and at the end of 3 and 6 months.

Results

Ninety-nine patients completed the 6-month follow-up (intervention group: n = 52; control

group: n = 47). On day 22, the fasting blood glucose (FBG) level of the intervention group

was lower than that of the control group (p < 0.05), and the control group’s SDS increased

significantly compared with the baseline value (p < 0.05). At the end of 3 months, glycated

hemoglobin (HbA1c) and FBG levels in the intervention group decreased significantly com-

pared with those in the control group (p < 0.01). At the end of 6 months, the intervention

group showed a significant decrease in postprandial blood glucose, triglyceride, and low-

density lipoprotein cholesterol levels as well as waist-to-hip ratio compared with the control

PLOS ONE

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 1 / 13

a1111111111

a1111111111

a1111111111

a1111111111

a1111111111

OPEN ACCESS

Citation: Yin W, Liu Y, Hu H, Sun J, Liu Y, Wang Z

(2022) Telemedicine management of type 2

diabetes mellitus in obese and overweight young

and middle-aged patients during COVID-19

outbreak: A single-center, prospective, randomized

control study. PLoS ONE 17(9): e0275251. https://

doi.org/10.1371/journal.pone.0275251

Editor: Ferdinando Carlo Sasso, University of

Campania Luigi Vanvitelli: Universita degli Studi

della Campania Luigi Vanvitelli, ITALY

Received: May 12, 2022

Accepted: September 12, 2022

Published: September 29, 2022

Peer Review History: PLOS recognizes the

benefits of transparency in the peer review

process; therefore, we enable the publication of

all of the content of peer review and author

responses alongside final, published articles. The

editorial history of this article is available here:

https://doi.org/10.1371/journal.pone.0275251

Copyright: © 2022 Yin et al. This is an open access

article distributed under the terms of the Creative

Commons Attribution License, which permits

unrestricted use, distribution, and reproduction in

any medium, provided the original author and

source are credited.

group (p < 0.05); moreover, the intervention group showed lower SDS scores than the base-

line value (p < 0.05). Further, the intervention group showed a significant reduction in BMI

compared with the control group at the end of 3 and 6 months (p < 0.01).

Conclusion

Telemedicine is a beneficial strategy for achieving remotely supervised blood glucose regu-

lation, weight loss, and depression relief in patients with T2DM.

Trial registration

ClinicalTrials.gov Identifier: NCT04723550.

Introduction

The rapid outbreak of coronavirus disease-2019 (COVID-19) adversely affected the daily life

of people worldwide. Due to the spread of the disease at a pandemic level, hospitals in China

implemented strict control measures, including limiting outpatient visits and inpatient admis-

sions as well as reducing operations to avoid cross-contamination caused by personnel move-

ment [1]. However, these epidemic prevention and control measures restricted the patients’

access to outpatient follow-up, blood glucose monitoring, and drug supply. Patients with dia-

betes mellitus were greatly affected during this period in terms of their self-efficacy and man-

agement ability [2]. Khare and Jindal followed up with 143 subjects who stayed at home for 3

months due to the nationwide lockdown and found that 56 (39.16%) had significantly elevated

blood glucose levels and required additional medication [3]. Verma et al. revealed that the

mean glycated hemoglobin (HbA1c) level of 52 patients during COVID-19 isolation (10% ± 1.5%) was significantly higher than their pre-pandemic mean value (8.8% + 1.3%) [4]. In addi-

tion, patients with diabetes are more likely to contract COVID-19 and exhibit a poor prognosis

[5–7]. Therefore, it is necessary to use the existing limited medical resources to ensure appro-

priate follow-up and management of patients with diabetes during the COVID-19 pandemic.

Telemedicine refers to remote diagnosis, treatment, and consultation of patients using

remote communication, digital holography, modern electronic technology, and computer

multimedia [8]. It allows the patient to avail full benefits of medical technology and equipment

available at higher medical centers as well as receive disease-related counseling and education

when social interaction or direct contact with the healthcare providers is not possible [9].

Arabi et al. has demonstrated that telemedicine had a positive influence on blood glucose con-

trol in patients with diabetes during the COVID-19 pandemic [10]. Notably, sudden lifestyle

changes that occurred during the COVID-19 lockdown in terms of restriction of outdoor

movement and social interaction had a great impact on the young and middle-aged popula-

tion, which may have led to physical and mental diseases, such as obesity, depression, and anx-

iety [11, 12]. Therefore, this study aimed to assess the effect of telemedicine on glycemic

control and diabetes-related anxiety symptoms in obese and overweight young and middle-

aged patients with type 2 diabetes mellitus (T2DM) during the COVID-19 pandemic.

Methods

Ethical considerations

This single-center, parallel-group randomized control study was conducted as per the “The

Code of Ethics of the World Medical Association (Declaration of Helsinki)” and was approved

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 2 / 13

Data Availability Statement: All data are available

from the https://clinicaltrials.gov/ (NCT number:

04723550).

Funding: This study was supported by grant no.

2020QN80 from the Fundanmental Research

Funds for the Central Universities. The funders had

no role in study design, data collection and

analysis, decision to publish, or preparation of the

manuscript.

Competing interests: The authors have declared

that no competing interests exist.

by the Ethics Committee of Affiliated Hospital of China University of Mining and Technology

(xyy11[2020]40). The patients consented to the procedure, and written informed consent was

obtained from them before enrollment. The trial was registered with ClinicalTrials

(NCT04723550).

Study patients and recruitment

In January 2021, the government imposed a strict lockdown due to the COVID-19 outbreak in

certain areas of China. For patients with T2DM who had visited our outpatient endocrine

clinic before the pandemic, we conducted a telephone follow-up interview. These patients

were considered candidates for this randomized controlled trial. Patients fulfilling the follow-

ing criteria were included in this study: physician diagnosis of T2DM for>6 months [13];

HbA1c level of 7.0%–10.0%; quarantined for 21 days due to COVID-19; aged 18–55 years;

body mass index (BMI)� 24 kg/m2; and can use smartphone and internet. The exclusion cri-

teria were as follows: insulin pump users; patients with a history of symptomatic cardiovascu-

lar disease (myocardial infarction, angina pectoris, surgical or endovascular intervention,

stroke older than 6 months, or symptomatic lower limb arteritis); pregnant or lactating

women or those who became pregnant during the study; patients who underwent obesity sur-

gery for >1 year; those diagnosed with COVID-19 infection; those with other comorbidities,

such as chronic heart disease, cerebrovascular disease, HIV/AIDS, cancer, emphysema,

chronic liver or kidney disease, which may affect the patients’ ability to follow the tailored

advice.

Patient randomization and data collection procedure

All patients were randomly assigned to the intervention or control group using a random

number sequence generated using the SPSS software (version 17.0; IBM Corp., Armonk, NY)

in batches of six patients. After enrollment, all patients underwent an initial physical examina-

tion and blood sample collection, followed by a mandatory home quarantine for 21 days. Dur-

ing the isolation period, the control group was followed up through telephone once a week.

1. Glucose data management: Postprandial blood glucose (PBG) and fasting blood glucose

(FBG) levels were monitored using a glucometer. Patients in the intervention group were

provided training for independently using the hospital’s telemedicine app. The glucometer

was connected to the patient’s mobile phone via Bluetooth. The glucometer data were then

automatically transferred to the hospital telemedicine app. The patients were followed up

four times a week (at least once during the weekends) in the first 3 months and twice a

week (at least once during the weekends) in the next 3 months. Doctors reminded patients

in the intervention group to monitor their blood glucose levels and provided medical advice

through the telemedicine system.

Patients in the control group were followed up through conventional outpatient clinic

appointments every 2 weeks, and telephone follow-up was used during the isolation period.

2. Diet guidance and exercise advice: For the intervention group, the dietitian advised the

patients on energy intake and food exchange methods. They were provided custom-tailored

dietary recommendations and were asked to consume the required calories and upload

their daily dietary intake on the telemedicine app. Additionally, the app recorded the

patients’ daily steps and automatically transferred them to the medical server. Further, exer-

cise guidance was provided to each patient in the intervention group.

The control group received traditional health education, which included diet, exercise, and

medication guidance, during clinic visits.

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 3 / 13

All patients were provided outpatient care and were followed up by the same medical team

at 22 days, 3 months, and 6 months after enrollment.

Outcome measurement

To assess the patients’ progress, the following data were reviewed four times (at baseline, 21

days after enrollment, and 3 and 6 months after enrollment): treatment, physical examination,

laboratory investigations, and Self-Rating Depression Scale (SDS) scores. Additionally, the

number of hypoglycemic episodes and the number of patients who dropped out of the study

due to any adverse event were calculated. The SDS is a self-reported 20-item questionnaire

assessing depressive state—each item is ranked on a 4-point response scale (1 indicates “a little

of the time” and 4 denotes “most of the time”) [14]. The total score ranges from 20 to 80, and

the reference value for normal SDS in adults is<50.

Sample size

A previous observational study involving 10 patients reported a decrease in the HbA1c levels

by 2.41% (standard deviation [SD] = 1.38%) in the outpatient group and 1.67% (SD = 1.15%)

in the telemedicine group after 6 months of follow-up. Based on these results, we considered a

bilateral α = 0.05 to achieve 80% test power and an equal ratio of patient allocation to the con-

trol and intervention groups (1:1) and computed a target sample size of 48 patients in each

group using PASS 15.0 (NCSS LLC). Further, accounting for a 20% loss of sample to follow-up

or refusal to follow-up, a final sample size of 60 patients in each group was considered.

Statistical analysis

All data were analyzed using the SPSS Statistics software (version 17.0; IBM Corp., Armonk,

NY). Normally distributed variables are expressed as means and SD, and non-normally dis-

tributed variables are expressed as median and interquartile range (IQR). Between-group dif-

ferences for normally distributed variables were assessed using an independent samples t-test,

whereas those for non-normally distributed variables were assessed using the Mann–Whitney

U test. For within-group comparisons, normally distributed variables were tested using a

paired t-test, whereas non-normally distributed variables were tested using the Wilcoxon

signed rank test. The figures were created using GraphPad software (GraphPad Prism version

8.4.2).

The primary outcome, HbA1c level, was evaluated at the four aforementioned time points

using two-way repeated-measures analysis of variance (ANOVA). The group effect, time effect,

and the effect of the interaction between group and time (group × time) were compared

between and within the groups. Bonferroni correction was used for post-hoc comparison. A p-

value of<0.05 was considered statistically significant for all analyses.

Results

Of the 201 potentially eligible patients, 120 expressed interest and were screened for suitability

for inclusion in this study. Of these, 6 and 11 patients in the intervention and control groups

voluntarily withdrew from the study, respectively, and two in each group were lost to follow-

up. A total of 47 and 52 patients in the control and intervention groups completed the

6-month follow-up, respectively (Fig 1). According to the COVID-19 diagnostic criteria, no

confirmed cases of COVID-19 were found in this study [15, 16]. The baseline characteristics of

the patients are summarized in Table 1. There were no significant between-group differences

in terms of age, physical findings, or biochemical indices.

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 4 / 13

Table 2 presents the summary statistics for outcome variables at the four follow-up time

points. Fig 2 shows a significant decrease in the HbA1c, FBG, PBG, triglyceride (TG), low-den-

sity lipoprotein cholesterol (LDL-C) levels as well as waist-to-hip ratio (WHR) and BMI in the

two groups at the end of 6 months, revealing a statistically significant difference from the base-

line values. The median FBG level in the intervention group was lower than the baseline value

(median [IQR] at 21 days: 6.52 [5.53–8.17] vs. baseline: 8.45 [7.69–9.35]; p < 0.01) and lower

than that in the control group (intervention: 6.52 [5.53–8.17] vs. control: 8.86 [7.77–9.77];

p< 0.01) at the end of 21-day isolation (Table 2, Fig 2B). Compared with the baseline, a statis-

tically significant decrease was observed in the HbA1c, FBG, PBG, TG, and LDL-C levels in

both groups and BMI in the intervention group at the end of 3 months (Table 2, Fig 2). Fur-

thermore, the improvement in the HbA1c and FBG levels in the intervention group was better

than that in the control group, and the difference was statistically significant (Table 2, Fig 2A

and 2B). At the end of the study (6-month follow-up), the extent of decrease in PBG, TG, and

LDL-C levels as well as WHR in the intervention group was larger and more significant than

that in the control group (Table 2, Fig 2).

The extent of reduction in BMI in the intervention group was significantly greater than in

the control group at the 3- and 6-month follow-ups (mean [SD] at 3-month follow-up: 27.10

[2.61] vs. baseline: 29.25 [2.93], p< 0.01; intervention group: 27.10 [2.61] vs. control group:

28.82 [2.57], p< 0.01) (6-month follow-up: 25.49 [2.35] vs. baseline: 29.25 [2.93], p< 0.01;

intervention group: 25.49 [2.35] vs. control group: 27.36 [1.90], p< 0.01) (Table 2, Fig 2H).

There was no statistically significant difference between the groups regarding blood pressure,

total cholesterol, high-density lipoprotein cholesterol, blood urea nitrogen, and creatinine lev-

els as well as estimated glomerular filtration rate at the 6-month follow-up.

Compared with the baseline, the mean SDS score in the control group increased after 21

days of isolation (44.02 [9.71] vs. 40.17 [11.60]; p< 0.05), whereas it decreased in the

Fig 1. Patient flow diagram.

https://doi.org/10.1371/journal.pone.0275251.g001

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 5 / 13

intervention group at the 6-month follow-up (37.08 [9.16] vs. 40.63 [11.10], p< 0.05). More-

over, the score in the intervention group was lower than that in the control group. These dif-

ferences showed a decreasing trend but were not statistically significant (Table 2, Fig 2G).

Trends noted in HbA1c levels during follow-up (Table 3)

The two-way repeated-measures ANOVA revealed that the main effect of the group was not

significant (F = 2.558, p = 0.113, partial η2 = 0.026), whereas time (F = 86.089, p< 0.001, par-

tial η2 = 0.470) and group × time (F = 3.498, p = 0.019, partial η2 = 0.035) showed significant

effects. The separate group effect was not significant at baseline (F = 0.106, p = 0.746, partial η2

= 0.001) and at the end of the isolation period (F = 0.361, p = 0.55, partial η2 = 0.004). How-

ever, at the end of 3 months, the simple effect of the group was significant (F = 5.765,

p = 0.018, partial η2 = 0.056), which eventually became nonsignificant at the end of 6 months

(F = 3.659, p = 0.059, partial η2 = 0.036). The simple effect of time was significant in the inter-

vention (F = 59.184, p< 0.001, partial η2 = 0.651) and control (F = 29.001, p< 0.001, partial

η2 = 0.478) groups. Post-hoc comparisons showed that HbA1c levels in the intervention group

were lower at the end of 3 and 6 months than at baseline and at the end of isolation (p< 0.05),

whereas those in the control group at the end of 6 months were lower than that at the end of 3

months and lower than those at the baseline and end of isolation (p< 0.05; Table 3).

Discussion

The COVID-19 pandemic has been an unprecedented global health concern affecting millions

worldwide. Patients with diabetes are more susceptible to COVID-19 and reportedly have a

Table 1. Baseline characteristics of the two groups.

Characteristic Control (n = 47) Intervention (n = 52) P-value

Age (years) 47.00 (42.00–51.00) 47.50 (43.00–51.00) 0.64

Gender, male, n (%) 20 (38) 20 (43) 0.68

Diabetes mellitus, duration in years 3.00 (2.00–5.00) 4.00 (2.00–6.00) 0.08

FBG (mmol/L) 8.95 (8.31–9.46) 8.45 (7.69–9.35) 0.10

PBG (mmol/L) 12.76 (11.63–14.22) 12.73 (11.95–13.79) 0.64

HbA1c (%) 8.50 (0.80) 8.56 (0.88) 0.75

Blood pressure systolic (mm Hg) 138.10 (125.20–144.85) 135.75 (127.43–147.90) 0.90

Blood pressure diastolic (mm Hg) 87.90 (80.45–92.10) 87.10 (79.80–92.10) 0.61

BMI (kg/m2) 29.05 (3.31) 29.25 (2.93) 0.76

Waist-to-hip ratio 0.94 (0.04) 0.96 (0.04) 0.11

TC (mmol/L) 5.03 (4.59–5.34) 4.78 (4.60–5.51) 0.85

TG (mmol/L) 1.99 (1.69–2.42) 1.98 (1.73–2.42) 0.85

LDL-C (mmol/L) 3.60 (0.28) 3.70 (0.28) 0.08

HDL-C (mmol/L) 1.30 (1.15–1.41) 1.20 (1.08–1.34) 0.06

BUN (mmol/L) 6.00 (5.50–6.40) 6.00 (5.40–6.40) 0.94

Cr (mmol/L) 64.10 (58.95–69.20) 62.65 (59.18–69.33) 0.82

e-GFR(ml/min) 90.07 (9.15) 91.53 (9.11) 0.43

SDS 40.17 (11.60) 40.63 (11.10) 0.81

Data are presented as the means (standard deviation) for the normally distributed variables, the median (interquartile range) for the non-normally distributed variables

or the number of participants (%). FBG: fasting blood glucose, PBG: postprandial blood glucose, HbA1c: glycated hemoglobin, BMI: Body Mass Index, TC: total

cholesterol, TG: triglyceride, LDL-C: low-density lipoprotein cholesterol, HDL-C: high-density lipoprotein cholesterol, BUN: blood urea nitrogen, Cr: creatinine, e-

GFR: estimated glomerular filtration rate, SDS: Self-Rating Depression Scale

https://doi.org/10.1371/journal.pone.0275251.t001

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 6 / 13

Table 2. The follow-up data of the two groups.

Characteristics 21 days 3 months 6 months

Control Intervention P vs.

control

Control Intervention P-

value

Control Intervention P-

value

HbA1c(%) 8.41 (1.21) 8.57 (1.43) 0.55 7.25 (1.78)b 6.60 (1.31)bc 0.04 6.66 (1.63)b 6.14 (1.05)b 0.13

Change vs. baseline −0.09 (1.52) 0.01 (1.88) −1.17 (1.71) −1.95 (1.71) −1.84 (1.55) −2.41 (1.38)

FBG (mmol/L) 8.86 (7.77–

9.77)

6.52 (5.53–

8.17)bd <0.01 6.40 (5.35–

8.50)b 5.45 (4.81–7.01)bc 0.02 5.99 (4.68–

7.67)b 5.58 (4.88–6.78)b 0.65

Change vs. baseline −0.17 (1.66) −1.88 (1.36) −2.08 (2.15) −2.57 (2.24) −2.83 (2.03) −2.74 (1.96)

PBG (mmol/L) 12.37 (11.52–

13.18)

12.07 (11.05–

13.80)

0.60 7.88 (6.97–

9.06)b 7.74 (6.91–8.62)b 0.37 7.58 (6.79–

9.19)b 6.78 (6.22–

7.70)bd <0.01

Change vs. baseline −0.53 (1.86) −0.56 (2.10) −4.25 (2.01) −4.88 (1.79) −4.86 (2.10) −5.52 (2.14)

BMI (kg/m2) 28.73 (2.97) 28.72 (2.61) 0.93 28.82 (2.57) 27.10 (2.61)bd <0.01 27.36 (1.90)b 25.49 (2.35)bd <0.01

Change vs. baseline −0.32 (4.81) −0.54 (3.85) −0.60 (4.41) −2.16 (3.65) −1.68 (3.95) −3.77 (3.38)

Waist-to-hip ratio 0.95 (0.04) 0.94 (0.04) 0.82 0.95 (0.04) 0.96 (0.05) 0.77 0.91 (0.09)a 0.86 (0.11)bc 0.03

Change vs. baseline 0.003 (0.06) −0.02 (0.06) 0.01 (0.07) −0.001 (0.06) −0.04 (0.09) −0.09 (0.10)

TG (mmol/L) 2.10 (1.83–

2.40)

1.96 (1.63–2.17) 0.10 1.65 (1.52–

2.20)a 1.63 (1.45–1.99)a 0.40 1.67 (1.51–

1.98)b 1.56 (1.43–1.83)bc 0.02

Change vs. baseline 0.01 (0.57) −0.11 (0.55) −0.20 (0.65) −0.25 (0.67) −0.26 (0.54) −0.39 (0.61)

LDL-C (mmol/L) 3.65 (0.33) 3.63 (0.34) 0.78 3.47 (0.41)a 3.49 (0.37)b 0.70 3.39 (0.50)b 3.03 (0.58)bd <0.01

Change vs. baseline 0.05 (0.39) −0.07 (0.41) −0.13 (0.41) −0.21 (0.45) −0.21 (0.50) −0.68 (0.66)

SDS 44.02 (9.71)a 41.19 (9.38) 0.10 39.02 (10.12) 40.08 (10.64) 0.48 39.94 (9.50) 37.08 (9.16)a 0.09

Change vs. baseline 3.85 (12.19) 0.56 (12.49) −1.15 (12.95) −0.56 (12.70) −0.23 (11.24) −3.56 (12.41)

Blood pressure systolic

(mm Hg)

134.70

(122.05–

144.25)

136.75 (125.50–

144.43)

0.48 130.50

(124.70–

141.90)

131.20 (122.70–

142.55)

0.97 137.70

(126.45–

143.65)

132.55 (125.68–

141.98)

0.99

Change vs. baseline −2.71 (16.44) −0.94 (15.89) −3.43 (15.50) −3.34 (16.21) −0.56 (14.67) −2.47 (14.53)

Blood pressure, diastolic

(mm Hg)

86.80 (80.80–

91.95)

87.75 (80.63–

93.33)

0.61 85.30 (80.40–

89.25)

85.90 (82.98–

91.68)

0.54 86.50 (82.20–

90.85)

85.50 (81.05–

91.45)

0.64

Change vs. baseline −0.20 (9.15) 1.20 (10.02) −1.76 (8.98) 1.13 (9.06) −0.52 (8.75) −0.11 (9.73)

TC (mmol/L) 4.89 (4.51–

5.44)

4.85 (4.62–5.26) 0.46 5.15 (4.64–

5.42)

5.00 (4.46–5.49) 0.34 4.96 (4.52–

5.42)

4.84 (4.57–5.51) 0.78

Change vs. baseline 0.0002 (0.66) −0.06 (0.71) 0.11 (0.71) 0.02 (0.72) −0.01 (0.74) 0.02 (0.72)

HDL-C (mmol/L) 1.23 (1.10–

1.34)

1.26 (1.12–1.39) 0.64 1.32 (1.14–

1.41)

1.26 (1.10–1.36) 0.27 1.20 (1.07–

1.37)

1.22 (1.12–1.38) 0.53

Change vs. baseline −0.04 (2.23) 0.03 (0.21) 0.0002 (0.22) 0.03 (0.22) −0.05 (0.25) 0.04 (0.23)

BUN (mmol/L) 5.70 (5.05–

6.40)

6.00 (5.35–6.60) 0.21 5.70 (5.40–

6.20)

5.80 (5.40–6.53) 0.43 5.90 (5.20–

6.10)

6.05 (5.18–6.50) 0.24

Change vs. baseline −0.17 (0.87) 0.02 (0.87) −0.12 (0.81) −0.01 (0.83) −0.16 (0.93) −0.01 (0.84)

Cr (mmol/L) 68.70 (62.7–

73.8)

67.20 (61.80–

74.10)

0.82 65.80 (57.45–

73.40)

67.15 (60.83–

73.40)

0.67 61.50 (56.65–

66.55)

67.85 (58.95–

6.50)

0.76

Change vs. baseline 2.57 (9.55) 2.32 (10.16) 0.34 (10.86) 1.83 (11.56) −2.80 (11.19) 1.77 (10.62)

e-GFR(ml/min) 91.95 (9.65) 91.30 (9.46) 0.70 89.84 (8.75) 89.93 (6.94) 0.95 90.39 (8.36) 89.97 (9.59) 0.82

Change vs. baseline 1.88 (13.82) −0.23 (14.84) −0.23 (12.98) −1.59 (13.21) 0.32 (13.93) −1.56 (13.11)

Data are presented as the means (SD) for the normally distributed variables, the median (IQR) for the non-normally distributed variables or the number of participants

(%). FBG: fasting blood glucose, PBG: postprandial blood glucose, HbA1c: glycated hemoglobin, BMI: Body Mass Index, TC: total cholesterol, TG: triglyceride, LDL-C:

low-density lipoprotein cholesterol, HDL-C: high-density lipoprotein cholesterol, BUN: blood urea nitrogen, Cr: creatinine, e-GFR: estimated glomerular filtration rate,

SDS: Self-Rating Depression Scale. In comparison with baseline, “a” indicates p < 0.05 versus baseline; “b” indicates p < 0.01 versus baseline; “�” indicates p < 0.05

versus control group; and “��” indicates p < 0.05 versus control group.

https://doi.org/10.1371/journal.pone.0275251.t002

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 7 / 13

rapid disease progression, high severity, and high fatality rate after infection [17]. The upregu-

lation of the angiotensin-converting enzyme (ACE)-2 receptor gene in the cardiomyocytes of

patients with diabetes mellitus along with nonenzymatic glycation may increase the suscepti-

bility to COVID-19 infection in patients with diabetes by promoting the entry of severe acute

respiratory syndrome coronavirus 2 into the cell [18]. Furthermore, the immune dysfunction,

proinflammatory cytokine environment, hypoglycemic state, and coagulopathy of patients

with diabetes contribute to a poor prognosis and complications of COVID-19 by increasing

the risk of mechanical ventilation, shock, and multiple organ failure, eventually leading to

death [19–23]. Sardu showed that early glycemic control reduces adverse events in hospitalized

patients with hyperglycemic COVID-19 with or without a previous diagnosis of diabetes [23].

Therefore, ensuring competent management of diabetes and taking scientific and effective

measures during the pandemic is crucial to improving immunity and reducing the risk of

infection in patients with diabetes.

Fig 2. Targets of the study. (A-H): Changes in the HbA1c, FBG, PBG, TG, and LDL-C levels as well as SDS, WHR, and BMI over 6 months of the study in the

control and intervention groups. HbA1c: glycated hemoglobin, FBG: fasting blood glucose, PBG: postprandial blood glucose, WHR: waist-to-hip ratio, TG:

triglyceride, LDL-C: low-density lipoprotein cholesterol, SDS: Self-Rating Depression Scale, BMI: body mass index. “a” indicates p< 0.05 vs. baseline; “b”

indicates p< 0.01 vs. baseline; “�” indicates p< 0.05 vs. control group; “��” indicates p< 0.05 vs. control group.

https://doi.org/10.1371/journal.pone.0275251.g002

Table 3. A comparison of the primary outcome, HbA1c, throughout the 6-month follow-up.

n Baseline 21 days 3 months 6 months F P Partial η2

Intervention 52 8.56 ± 0.88 8.57 ± 1.43 6.60 ± 1.31ab 6.14 ± 1.05ab

Control 47 8.50 ± 0.80 8.41 ± 1.21 7.34 ± 1.71ab 6.66 ± 1.63abc

Group main effect 2.558 0.113 0.026

Time main effect 86.089 <0.001 0.470

Group�Time interaction effect 3.498 0.019 0.035

Compared with baseline, “a” indicates p < 0.05; compared with 21 days, “b” indicates p < 0.05; compared with the end of 3 months, “c” indicates p < 0.05.

https://doi.org/10.1371/journal.pone.0275251.t003

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 8 / 13

However, the stringent pandemic control measures have posed new challenges to the man-

agement and follow-up of patients with chronic illnesses. In this regard, telemedicine offers

patients an opportunity for remote diagnosis, treatment, and consultation; reduces medical

expenses; and prevents cross-infection during outpatient visits. Few studies have indicated

that telemedicine should be actively used to control the development of chronic illnesses dur-

ing an epidemic [24]. Moreover, Bahl et al. suggested that telemedicine should be explicitly

promoted as an alternative to traditional medical care during the COVID-19 pandemic, partic-

ularly during isolation [25]. Currently, several studies on telemedicine management of diabe-

tes have reported favorable outcomes; however, these studies mostly comprised older patients.

There has been a recent increase in the incidence of T2DM in younger patients who belong to

the age group that was considerably affected by the lockdown measures. Therefore, we assessed

the efficacy of telemedicine in obese and overweight young and middle-aged patients with dia-

betes. Although obesity is not the only risk factor for T2DM, it can significantly increase the

risk of complications in patients with T2DM [26]. Therefore, we focused on the application of

telemedicine in the glycemic control of obese and overweight patients with T2DM during and

after a 21-day isolation period.

In our study, the intervention and control groups managed blood glucose levels through

telemedicine and outpatient follow-up, respectively. At the end of the isolation period, FBG

level in the intervention group was lower than that in the control group and the baseline value.

After 3 months, both groups showed significant reductions in HbA1c and FBG levels, but the

reduction was greater in the intervention group. Likewise, at the end of 6 months, PBG level

decreased more significantly in the intervention group than in the control group. Additionally,

10 and 6 patients in the control and intervention groups, respectively, required adjustment of

drug dosage for titration to achieve glycemic control during the study. At the end of 6 months,

the rate of discontinuation of antidiabetic medications was 11.5% in the intervention group

versus 2.1% in the control group. None of the patients in either group experienced any serious

adverse events or aggravation of complications. Further, the intervention group showed signif-

icantly fewer hypoglycemic events than the control group (n = 8 vs. n = 13). This difference

may be attributed to the fact that the medical team could identify the risk of hypoglycemia at

an appropriate time in the intervention group through telemedicine and take appropriate mea-

sures. These results revealed the positive effect of telemedicine in ensuring adequate glycemic

control in patients with T2DM and suggested that the use of telemedicine should be promoted

and popularized, especially when diligent physical monitoring is not possible.

Obesity and T2DM are closely associated and are growing global public health problems.

Weight gain is an independent risk factor for T2DM, and 87.5% of adults with diabetes are

overweight or obese [27]. The recent COVID-19 isolation measures have further reduced

physical activity and promoted sedentary behavior. Renzo et al. studied the eating habits of

and lifestyle changes in 3,533 respondents in Italy during the COVID-19 pandemic and

reported that weight gain was observed in 48.6% of the study population [28]. These results

highlight the fact that weight control is a challenging task and is particularly important in man-

aging diabetes during the COVID-19 epidemic. In our study, at the end of 3 and 6 months,

BMI was significantly lower in the intervention group than in the control group; similarly, at

the end of 6 months, WHR reduced significantly in the intervention group compared with the

control group. This indicates that diabetes management through telemedicine can effectively

help in weight control and abdominal obesity improvement. Both groups received diet and

exercise instructions; however, the medical team, including doctors, nurses, and nutritionists,

could monitor the diet of patients in the intervention group more frequently through telemed-

icine, allowing for more precise guidance. Strategies such as weighing frequently, recording

energy intake, and tracking physical activity are currently used to improve compliance;

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 9 / 13

however, these strategies are inadequate to achieve significant weight loss without feedback

[29]. Weight and blood glucose management can be more effective by obtaining timely feed-

back from patients. In this study, we provided individualized, simple, and feasible home exer-

cise guidance according to the actual energy intake of different patients. The majority of

patients (45/52) could complete most of the prescribed exercise at the end of the study and

showed significant weight loss. In addition, the fast food consumed by young and middle-aged

people during their daily work is often calorie-rich junk food. During the 21-day isolation and

under expert dietary guidance, the intervention group could improve their dietary structure,

which may also be one of the reasons for the efficient glycemic and weight control in these

patients.

Depression is an independent risk factor for T2DM [30]. Depression decreases dietary and

treatment adherence in patients with diabetes, thereby affecting glycemic control, aggravating

symptoms, and reducing their quality of life [31, 32]. According to a meta-analysis of 11 longi-

tudinal studies, depression can considerably increase the risk of developing T2DM by 37%

[33]. Further, several studies have shown that the incidence of depression in patients with dia-

betes is significantly higher than that in normal individuals. Khaledi et al. reported that

approximately 25% of adults with T2DM suffer from depression [34]. Zhong collected the

demographic and lifestyle data of 19,802 participants from 34 provinces in China during the

COVID-19 pandemic through a web-based survey and revealed that participants with chronic

illnesses had a 93% increased risk of high anxiety levels [35]. Likewise, Xiang et al. highlighted

the need for urgent interventions for the mental health of patients with chronic illnesses dur-

ing the COVID-19 pandemic [36]. In the current study, we assessed the effect of telemedicine

in addition to glycemic and weight control on diabetes-related anxiety symptoms in obese and

overweight young and middle-aged patients with T2DM. We observed that after the 21-day

isolation period, the SDS score of the control group was significantly higher than the baseline

value (p< 0.05). After the 6-month follow-up, the SDS score of the intervention group signifi-

cantly decreased compared with the baseline value (p< 0.05) and was lower than that of the

control group, although the difference was not statistically significant. Based on these results,

we can infer that depressive symptoms in patients with chronic illnesses could be relieved to

some extent via psychological counseling and communication through telemedicine. Addi-

tionally, the favorable results for telemedicine could be attributed to the fact that the study

population included young and middle-aged patients who have higher acceptance of modern

communication devices such as smartphones than other age groups.

Recently, the role of multifaceted management of patients with T2DM has been repeatedly

emphasized [37]. The telemedicine measures provided in this study focused more on the pop-

ularization of diabetes mellitus-related knowledge, diet/exercise guidance, blood glucose mon-

itoring, and adjustment of glucose-lowering regimen. We believe that patients should receive

more comprehensive multifactorial control, if conditions permit. However, further research is

warranted in this regard.

This study has some limitations. First, it is a single-center study. Second, the study popula-

tion was limited to individuals who could use mobile phones and internet independently,

which may have affected the applicability of our results to patients with T2DM who require

constant monitoring and cannot access modern technology. Third, the authors must acknowl-

edge the shortcomings of the missing data in this study, in which 120 patients were enrolled

and only 99 completed the 6-month follow-up. Due to these missing data (17.5% of the study

population), the findings of this study should be considered with caution.

In conclusion, this study revealed that telemedicine had a positive effect on glycemic con-

trol and weight management in obese and overweight young and middle-aged patients with

T2DM during the COVID-19 pandemic. Moreover, follow-up through telemedicine was

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 10 / 13

superior to that through conventional outpatient clinic appointment in controlling depressive

symptoms during the pandemic. These results are extremely encouraging and corroborate the

importance of telehealth measures to provide more professional, safe, economic, and personal-

ized management and improve the ability of patients to self-manage their conditions, resulting

in a higher quality of life. Furthermore, this study provides a basis for the development of an

individualized telemedicine management model for diabetes.

Supporting information

S1 Checklist. CONSORT 2010 checklist of information to include when reporting a rando-

mised trial�.

(DOC)

S1 Protocol.

(PDF)

Acknowledgments

We thank the participants who took part in the randomized controlled trial.

Author Contributions

Conceptualization: Wenwen Yin.

Data curation: Wenwen Yin, Yawen Liu, Hao Hu.

Formal analysis: Wenwen Yin, Yawen Liu.

Investigation: Wenwen Yin, Hao Hu, Yuanyuan Liu.

Methodology: Wenwen Yin, Jin Sun.

Project administration: Yawen Liu.

Supervision: Yawen Liu, Zhaoling Wang.

Writing – original draft: Wenwen Yin.

Writing – review & editing: Zhaoling Wang.

References 1. Yu X, Li N, Dong Y. Observation on China’s Strategies to Prevent the Resurgence of the COVID-19 Epi-

demic. Risk Manag Healthc Policy. 2021; 14:2011–9. Epub 2021/05/28. https://doi.org/10.2147/RMHP.

S305413 PMID: 34040464; PubMed Central PMCID: PMC8140918.

2. Paulsamy P, Ashraf R, Alshahrani SH, Periannan K, Qureshi AA, Venkatesan K, et al. Social Support,

Self-Care Behaviour and Self-Efficacy in Patients with Type 2 Diabetes during the COVID-19 Pan-

demic: A Cross-Sectional Study. Healthcare (Basel). 2021; 9(11). Epub 2021/11/28. https://doi.org/10.

3390/healthcare9111607 PMID: 34828652; PubMed Central PMCID: PMC8622453.

3. Khare J, Jindal S. Observational study on Effect of Lock Down due to COVID 19 on glycemic control in

patients with Diabetes: Experience from Central India. Diabetes Metab Syndr. 2020; 14(6):1571–4.

Epub 2020/08/29. https://doi.org/10.1016/j.dsx.2020.08.012 PMID: 32858474; PubMed Central

PMCID: PMC7438221.

4. Verma A, Rajput R, Verma S, Balania VKB, Jangra B. Impact of lockdown in COVID 19 on glycemic

control in patients with type 1 Diabetes Mellitus. Diabetes & Metabolic Syndrome: Clinical Research &

Reviews. 2020; 14(5):1213–6. https://doi.org/10.1016/j.dsx.2020.07.016 PMID: 32679527

5. Guo W, Li M, Dong Y, Zhou H, Zhang Z, Tian C, et al. Diabetes is a risk factor for the progression and

prognosis of COVID-19. Diabetes Metab Res Rev. 2020; 36(7):e3319. Epub 2020/04/02. https://doi.

org/10.1002/dmrr.3319 PMID: 32233013; PubMed Central PMCID: PMC7228407.

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 11 / 13

6. Cuschieri S, Grech S. COVID-19 and diabetes: The why, the what and the how. J Diabetes Complica-

tions. 2020; 34(9):107637. Epub 2020/05/28. https://doi.org/10.1016/j.jdiacomp.2020.107637 PMID:

32456846; PubMed Central PMCID: PMC7242955.

7. Lim S, Bae JH, Kwon HS, Nauck MA. COVID-19 and diabetes mellitus: from pathophysiology to clinical

management. Nat Rev Endocrinol. 2021; 17(1):11–30. Epub 2020/11/15. https://doi.org/10.1038/

s41574-020-00435-4 PMID: 33188364; PubMed Central PMCID: PMC7664589

8. Cobry EC, Wadwa RP. The future of telehealth in type 1 diabetes. Curr Opin Endocrinol Diabetes Obes.

2022; 29(4):397–402. Epub 2022/07/02. https://doi.org/10.1097/MED.0000000000000745 PMID:

35777972.

9. Galiero R, Pafundi PC, Nevola R, Rinaldi L, Acierno C, Caturano A, et al. The Importance of Telemedi-

cine during COVID-19 Pandemic: A Focus on Diabetic Retinopathy. J Diabetes Res. 2020;

2020:9036847. Epub 2020/10/31. https://doi.org/10.1155/2020/9036847 PMID: 33123599; PubMed

Central PMCID: PMC7584941 research, authorship, and/or publication of this article.

10. A MT, T JA, Rsheed AMB, Alrasheedy AA, W AL, F AL, et al. The impact of telemedicine on patients

with uncontrolled type 2 diabetes mellitus during the COVID-19 pandemic in Saudi Arabia: Findings and

implications. J Telemed Telecare. 2021:1357633X20985763. Epub 2021/02/03. https://doi.org/10.

1177/1357633X20985763 PMID: 33525952.

11. Delmastro M, Zamariola G. Depressive symptoms in response to COVID-19 and lockdown: a cross-

sectional study on the Italian population. Sci Rep. 2020; 10(1):22457–. https://doi.org/10.1038/s41598-

020-79850-6 PMID: 33384427.

12. Dragun R, Veček NN, MarendićM, Pribisalić A, ivićG, Cena H, et al. Have Lifestyle Habits and Psycho-

logical Well-Being Changed among Adolescents and Medical Students Due to COVID-19 Lockdown in

Croatia? Nutrients. 2020; 13(1):97. https://doi.org/10.3390/nu13010097 PMID: 33396690.

13. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complica-

tions. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation.

Diabet Med. 1998; 15(7):539–53. Epub 1998/08/01. https://doi.org/10.1002/(SICI)1096-9136(199807)

15:7<539::AID-DIA668>3.0.CO;2-S PMID: 9686693.

14. ZUNG WWK. A Self-Rating Depression Scale. Archives of General Psychiatry. 1965; 12(1):63–70.

https://doi.org/10.1001/archpsyc.1965.01720310065008 PMID: 14221692

15. Corman VM, Eckerle I, Bleicker T, Zaki A, Landt O, Eschbach-Bludau M, et al. Detection of a novel

human coronavirus by real-time reverse-transcription polymerase chain reaction. Euro Surveill. 2012;

17(39). Epub 2012/10/09. https://doi.org/10.2807/ese.17.39.20285-en PMID: 23041020.

16. Huang P, Wang H, Cao Z, Jin H, Chi H, Zhao J, et al. A Rapid and Specific Assay for the Detection of

MERS-CoV. Front Microbiol. 2018; 9:1101. Epub 2018/06/14. https://doi.org/10.3389/fmicb.2018.

01101 PMID: 29896174; PubMed Central PMCID: PMC5987675.

17. Gupta R, Ghosh A, Singh AK, Misra A. Clinical considerations for patients with diabetes in times of

COVID-19 epidemic. Diabetes & metabolic syndrome. 2020; 14(3):211.

18. D’Onofrio N, Scisciola L, Sardu C, Trotta MC, De Feo M, Maiello C, et al. Glycated ACE2 receptor in dia-

betes: open door for SARS-COV-2 entry in cardiomyocyte. Cardiovasc Diabetol. 2021; 20(1):99. Epub

2021/05/09. https://doi.org/10.1186/s12933-021-01286-7 PMID: 33962629; PubMed Central PMCID:

PMC8104461.

19. Frydrych LM, Bian G, O’Lone DE, Ward PA, Delano MJ. Obesity and type 2 diabetes mellitus drive

immune dysfunction, infection development, and sepsis mortality. J Leukoc Biol. 2018; 104(3):525–34.

Epub 2018/08/02. https://doi.org/10.1002/JLB.5VMR0118-021RR PMID: 30066958.

20. Pal R, Bhadada SK. COVID-19 and diabetes mellitus: An unholy interaction of two pandemics. Diabetes

Metab Syndr. 2020; 14(4):513–7. Epub 2020/05/11. https://doi.org/10.1016/j.dsx.2020.04.049 PMID:

32388331; PubMed Central PMCID: PMC7202837.

21. Zhou J, Tan J. Diabetes patients with COVID-19 need better blood glucose management in Wuhan,

China. Metabolism. 2020; 107:154216. Epub 2020/03/30. https://doi.org/10.1016/j.metabol.2020.

154216 PMID: 32220612; PubMed Central PMCID: PMC7102634 conflict of interest.

22. Sardu C, Gargiulo G, Esposito G, Paolisso G, Marfella R. Impact of diabetes mellitus on clinical out-

comes in patients affected by Covid-19. Cardiovasc Diabetol. 2020; 19(1):76. Epub 2020/06/13. https://

doi.org/10.1186/s12933-020-01047-y PMID: 32527257; PubMed Central PMCID: PMC7289072.

23. Sardu C, D’Onofrio N, Balestrieri ML, Barbieri M, Rizzo MR, Messina V, et al. Hyperglycaemia on

admission to hospital and COVID-19. Diabetologia. 2020; 63(11):2486–7. Epub 2020/07/08. https://doi.

org/10.1007/s00125-020-05216-2 PMID: 32632527; PubMed Central PMCID: PMC7335920.

24. Mader JK. Personal Experiences With Coronavirus Disease 2019 and Diabetes: The Time for Telemed-

icine is Now. J Diabetes Sci Technol. 2020; 14(4):752–3. Epub 2020/05/24. https://doi.org/10.1177/

1932296820930289 PMID: 32443942; PubMed Central PMCID: PMC7673167.

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 12 / 13

25. Bahl S, Singh RP, Javaid M, Khan IH, Vaishya R, Suman R. Telemedicine technologies for confronting

COVID-19 pandemic: a review. Journal of Industrial Integration and Management. 2020; 5(04):547–61.

26. Bhupathiraju SN, Hu FB. Epidemiology of obesity and diabetes and their cardiovascular complications.

Circulation research. 2016; 118(11):1723–35. https://doi.org/10.1161/CIRCRESAHA.115.306825

PMID: 27230638

27. Afshin A, Forouzanfar MH, Reitsma MB, Sur P, Estep K, Lee A, et al. Health Effects of Overweight and

Obesity in 195 Countries over 25 Years. N Engl J Med. 2017; 377(1):13–27. Epub 2017/06/13. https://

doi.org/10.1056/NEJMoa1614362 PMID: 28604169; PubMed Central PMCID: PMC5477817.

28. Di Renzo L, Gualtieri P, Pivari F, Soldati L, Attinà A, Cinelli G, et al. Eating habits and lifestyle changes

during COVID-19 lockdown: an Italian survey. J Transl Med. 2020; 18(1):229. Epub 2020/06/10. https://

doi.org/10.1186/s12967-020-02399-5 PMID: 32513197; PubMed Central PMCID: PMC7278251.

29. Alencar M, Johnson K, Gray V, Mullur R, Gutierrez E, Dionico P. Telehealth-Based Health Coaching

Increases m-Health Device Adherence and Rate of Weight Loss in Obese Participants. Telemed J E

Health. 2020; 26(3):365–8. Epub 2019/04/18. https://doi.org/10.1089/tmj.2019.0017 PMID: 30994410;

PubMed Central PMCID: PMC7071022.

30. Hackett RA, Steptoe A. Type 2 diabetes mellitus and psychological stress—a modifiable risk factor. Nat

Rev Endocrinol. 2017; 13(9):547–60. Epub 2017/07/01. https://doi.org/10.1038/nrendo.2017.64 PMID:

28664919.

31. Ghosh A, Gupta R, Misra A. Telemedicine for diabetes care in India during COVID19 pandemic and

national lockdown period: Guidelines for physicians. Diabetes Metab Syndr. 2020; 14(4):273–6. Epub

2020/04/14. https://doi.org/10.1016/j.dsx.2020.04.001 PMID: 32283497; PubMed Central PMCID:

PMC7129346.

32. Joshi R, Atal S, Fatima Z, Balakrishnan S, Sharma S, Joshi A. Diabetes care during COVID-19 lock-

down at a tertiary care centre in India. Diabetes Res Clin Pract. 2020; 166:108316. Epub 2020/07/17.

https://doi.org/10.1016/j.diabres.2020.108316 PMID: 32673697; PubMed Central PMCID:

PMC7358163.

33. Knol MJ, Twisk JWR, Beekman ATF, Heine RJ, Snoek FJ, Pouwer F. Depression as a risk factor for the

onset of type 2 diabetes mellitus. A meta-analysis. Diabetologia. 2006; 49(5):837. https://doi.org/10.

1007/s00125-006-0159-x PMID: 16520921

34. Khaledi M, Haghighatdoost F, Feizi A, Aminorroaya A. The prevalence of comorbid depression in

patients with type 2 diabetes: an updated systematic review and meta-analysis on huge number of

observational studies. Acta Diabetologica. 2019; 56(6):631–50. https://doi.org/10.1007/s00592-019-

01295-9 PMID: 30903433

35. Zhong J, Zhong C, Qiu L, Li J, Lai J, Lu W, et al. Risk and protective factors for anxiety during COVID-

19 pandemic. BMC Public Health. 2021; 21(1):1063. Epub 2021/06/06. https://doi.org/10.1186/s12889-

021-11118-8 PMID: 34088290; PubMed Central PMCID: PMC8177261.

36. Xiang YT, Yang Y, Li W, Zhang L, Zhang Q, Cheung T, et al. Timely mental health care for the 2019

novel coronavirus outbreak is urgently needed. Lancet Psychiatry. 2020; 7(3):228–9. Epub 2020/02/08.

https://doi.org/10.1016/S2215-0366(20)30046-8 PMID: 32032543; PubMed Central PMCID:

PMC7128153.

37. Sasso FC, Pafundi PC, Simeon V, De Nicola L, Chiodini P, Galiero R, et al. Efficacy and durability of

multifactorial intervention on mortality and MACEs: a randomized clinical trial in type-2 diabetic kidney

disease. Cardiovasc Diabetol. 2021; 20(1):145. Epub 2021/07/18. https://doi.org/10.1186/s12933-021-

01343-1 PMID: 34271948; PubMed Central PMCID: PMC8285851.

PLOS ONE Telemedicine for T2DM during COVID-19 pandemic

PLOS ONE | https://doi.org/10.1371/journal.pone.0275251 September 29, 2022 13 / 13

Effect of a Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Patients With Persistently Poor Type 2 Diabetes Control A Randomized Clinical Trial Matthew J. Crowley, MD, MHS; Phillip E. Tarkington, MD; Hayden B. Bosworth, PhD; Amy S. Jeffreys, MStat; Cynthia J. Coffman, PhD; Matthew L. Maciejewski, PhD; Karen Steinhauser, PhD; Valerie A. Smith, DrPH; Moahad S. Dar, MD; Sonja K. Fredrickson, MD; Amy C. Mundy, NP; Elizabeth M. Strawbridge, RD; Teresa J. Marcano, MSN, RN; Donna L. Overby, RN; Nadya T. Majette Elliott, MPH; Susanne Danus, BS; David Edelman, MD

IMPORTANCE Persistently poorly controlled type 2 diabetes (PPDM) is common and causes poor outcomes. Comprehensive telehealth interventions could help address PPDM, but effectiveness is uncertain, and barriers impede use in clinical practice.

OBJECTIVE To address evidence gaps preventing use of comprehensive telehealth for PPDM by comparing a practical, comprehensive telehealth intervention to a simpler telehealth approach.

DESIGN, SETTING, AND PARTICIPANTS This active-comparator, parallel-arm, randomized clinical trial was conducted in 2 Veterans Affairs health care systems. From December 2018 to January 2020, 1128 outpatients with PPDM were assessed for eligibility and 200 were randomized; PPDM was defined as maintenance of hemoglobin A1c (HbA1c) level of 8.5% or higher for 1 year or longer despite engagement with clinic-based primary care and/or diabetes specialty care. Data analyses were preformed between March 2021 and May 2022.

INTERVENTIONS Each 12-month intervention was nurse-delivered and used only clinical staffing/resources. The comprehensive telehealth group (n = 101) received telemonitoring, self-management support, diet/activity support, medication management, and depression support. Patients assigned to the simpler intervention (n = 99) received telemonitoring and care coordination.

MAIN OUTCOMES AND MEASURES Primary (HbA1c) and secondary outcomes (diabetes distress, diabetes self-care, self-efficacy, body mass index, depression symptoms) were analyzed over 12 months using intent-to-treat linear mixed longitudinal models. Sensitivity analyses with multiple imputation and inclusion of clinical data examined the impact of missing HbA1c measurements. Adverse events and intervention costs were examined.

RESULTS The population (n = 200) had a mean (SD) age of 57.8 (8.2) years; 45 (22.5%) were women, 144 (72.0%) were of Black race, and 11 (5.5%) were of Hispanic/Latinx ethnicity. From baseline to 12 months, HbA1c change was −1.59% (10.17% to 8.58%) in the comprehensive telehealth group and −0.98% (10.17% to 9.19%) in the telemonitoring/care coordination group, for an estimated mean difference of −0.61% (95% CI, −1.12% to −0.11%; P = .02). Sensitivity analyses showed similar results. At 12 months, patients receiving comprehensive telehealth had significantly greater improvements in diabetes distress, diabetes self-care, and self-efficacy; no differences in body mass index or depression were seen. Adverse events were similar between groups. Comprehensive telehealth cost an additional $1519 per patient per year to deliver.

CONCLUSIONS AND RELEVANCE This randomized clinical trial found that compared with telemonitoring/care coordination, comprehensive telehealth improved multiple outcomes in patients with PPDM at a reasonable additional cost. This study supports consideration of comprehensive telehealth implementation for PPDM in systems with appropriate infrastructure and may enhance the value of telehealth during the COVID-19 pandemic and beyond.

TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT03520413

JAMA Intern Med. doi:10.1001/jamainternmed.2022.2947 Published online July 25, 2022.

Visual Abstract

Supplemental content

Author Affiliations: Author affiliations are listed at the end of this article.

Corresponding Author: Matthew J. Crowley, MD, MHS, Division of Endocrinology, Department of Medicine, Duke University School of Medicine; Durham VA Center of Innovation to Accelerate Discovery and Practice Transformation, Durham VAMC HSR&D (152), 508 Fulton St, Durham, NC 27705 (matthew. [email protected]).

Research

JAMA Internal Medicine | Original Investigation

(Reprinted) E1

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

P atients with persistently poor type 2 diabetes (T2D) con- trol disproportionately experience negative outcomes.1-3

We have defined persistently poorly controlled diabe- tes (PPDM) as maintenance of hemoglobin A1c (HbA1c) level of 8.5% or greater for more than 1 year despite receiving clinic- based diabetes care; 10% to 15% of all patients with T2D meet PPDM criteria.4,5 Given their high risk for complications and costs,1-3 patients with PPDM represent a compelling popula- tion for care delivery redesign.

Drivers of PPDM, including unavailable blood glucose data, medication nonadherence, suboptimal diet/activity, com- plex medications, and depression,6-12 can be difficult to ad- dress in the clinic setting.13,14 By facilitating contact outside of clinic, telehealth could improve outcomes in PPDM. Tele- health strategies targeting individual factors underlying poor T2D control reduce HbA1c level vs clinic-based care by 0.3% to 0.6%.15-19 Although such HbA1c changes may not suffice for patients with PPDM, combining multiple strategies into com- prehensive telehealth interventions could produce greater im- provement. However, multicomponent T2D interventions have achieved variable results.20-25

Beyond this uncertain effectiveness, other barriers have hindered implementation of comprehensive telehealth for PPDM in practice. Implementation barriers include interven- tion designs reliant on research-funded staff and resources, in- sufficient electronic health record (EHR) integration of pa- tient data, and uncertain reimbursement.26-29 Before comprehensive telehealth can become a real-world solution for PPDM, approaches are needed that are unambiguously ef- fective and also explicitly designed for feasible implementa- tion. The upsurge in telehealth use during the COVID-19 pan- demic has only strengthened the case for considering comprehensive telehealth as a means to address PPDM.30

We sought to address barriers to practical use of compre- hensive telehealth for PPDM by evaluating a comprehensive telehealth intervention in a randomized clinical trial (RCT). This intervention combined 5 strategies targeting contributors to PPDM: telemonitoring, self-management support, diet/ activity support, medication management, and depression sup- port. To facilitate eventual implementation, we explicitly de- signed the intervention for delivery by clinical Veterans Health Administration (VHA) Home Telehealth (HT) nurses using ex- isting clinical resources.

Methods The protocol for this active-comparator, parallel-arm RCT (NCT03520413) has been published5 and appears in Supplement 1. We compared a practical, comprehensive telehealth intervention with a simpler telehealth approach, consisting of telemonitoring and care coordination, which is already available in VHA practice for T2D. A usual-care comparator was deemed inappropriate because clinic-based care is by definition insufficient for PPDM.31

This RCT was conducted at 2 VHA sites (Durham, North Carolina, and Richmond, Virginia). Institutional review boards at both sites approved the study. This study followed the Con-

solidated Standards of Reporting Trials (CONSORT) reporting guideline.

Population We recruited patients with PPDM, defined as the following: di- agnosed T2D (International Statistical Classification of Dis- eases and Related Health Problems, Tenth Revision E11); 2 or more HbA1c values of 8.5% or greater during the prior year with none less than 8.5%; and at least 1 appointment during the prior year with a primary care clinician or diabetes specialist (en- docrinologist or other diabetes clinician). Exclusion criteria in- cluded the following: refusal to enroll in VHA HT (because both study interventions were delivered by HT nurses); factors mak- ing HbA1c reduction potentially inadvisable (age >70 years, metastatic cancer/life expectancy <5 years, recent cardiovas- cular disease complications, or prior hypoglycemic seizure/ coma); lack of telephone access; dementia, psychosis, or sub- stance use disorder; pregnancy; receiving dialysis or skilled nursing care; insulin pump use; or continuous glucose moni- tor use (unless also willing to submit self-monitored blood glu- cose [SMBG] data per HT protocol).

Race was determined by self-report and categorized as Asian, American Indian or Alaska Native, Black or African American, Native Hawaiian or Other Pacific Islander, White, other, or unknown. Because of low numbers in the other cat- egories, race was ultimately presented as Black or African American, White, or other race. Self-reported ethnicity was as- sessed using a single question: “Are you of Latino/a or His- panic origin or descent?” Race and ethnicity data were col- lected to facilitate generalizability assessment and to determine whether intervention effectiveness varied by these factors.

Recruitment and Enrollment After EHR screening, a research assistant mailed opt-out let- ters to potential participants, then conducted phone screen- ing. Eligible participants provided informed consent and un- derwent in-person baseline assessment; consented patients with a baseline HbA1c level of less than 8.5% were excluded

Key Points Question Compared with a simpler telehealth approach (telemonitoring and care coordination), can a practical, comprehensive telehealth intervention improve outcomes among patients whose type 2 diabetes remains persistently poorly controlled despite clinic-based care?

Findings In this randomized clinical trial of 200 adults with persistently poorly controlled type 2 diabetes, hemoglobin A1c

level improved by 1.59% at 12 months among those randomized to receive the comprehensive telehealth intervention, compared with 0.98% for the telemonitoring/care coordination group.

Meaning A comprehensive telehealth intervention improved outcomes in persistently poorly controlled type 2 diabetes compared with a simpler telehealth intervention; because it was explicitly designed for feasible use in clinical practice, this approach may warrant implementation in systems that need to improve diabetes control in which the requisite infrastructure is available.

Research Original Investigation Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes

E2 JAMA Internal Medicine Published online July 25, 2022 (Reprinted) jamainternalmedicine.com

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

before randomization. Given the high proportion of men in the VHA population, we oversampled women, aiming to achieve greater than 20% in the randomized population.

Randomization and Blinding Randomization was stratified with blocks of 2; stratification variables were site, prior VHA HT use, and preenrollment dia- betes specialty care (endocrinologist or other diabetes clini- cian). The computer-generated randomization sequence was accessible only to study statisticians. Patients received ran- domization assignments by phone from the project coordina- tor within 1 week of consent. Because participants received in- formation about both interventions during consent, they were not blinded to randomization arm. Research assistants col- lecting outcome data were blinded to participant randomiza- tion status.

Interventions Both 12-month interventions were delivered by clinical HT nurses rather than research staff. Although experienced with telehealth-based disease care, these nurses had no special- ized diabetes training. Individual nurses delivered only 1 study intervention, with no crossover. Each intervention was deliv- ered by 1 nurse in Durham, while in Richmond, 2 nurses de- livered comprehensive telehealth and 5 delivered telemoni- toring/care coordination.

Following randomization, all participants enrolled in VHA HT. Patients enrolled in HT used a telehealth device (Medtronic), blood glucose meter (Abbott), and connector cable; once connected to the blood glucose meter, the tele- health device automatically transmitted SMBG data to HT. Af- ter HT enrollment, participants began their assigned interven- tion; both groups also continued care with existing clinicians. Patients’ HbA1c goals were individualized per American Dia- betes Association guidelines.32

Comprehensive Telehealth Intervention This intervention comprised 5 nurse-delivered components (Figure 1)5: telemonitoring, self-management support, diet/ activity support (together with a study dietitian), medication management (together with a study medication manager), and depression support (together with a study psychiatrist). The

study dietitian, medication managers, and psychiatrists were clinicians, not research staff. Intervention nurses completed a single training session and received a manual. Nurses deliv- ered the intervention to participants during 26 every-2-week telephone encounters, which the nurses scheduled directly with participants; nurses could additionally be reached for acute issues. Clinical information was tracked using tem- plated EHR notes.

For the telemonitoring component, participants transmit- ted SMBG data up to 4 times daily based on their medication regimens but could monitor less frequently per nurse discre- tion. During each of the 26 scheduled encounters, the nurse reviewed SMBG data, reconciled medications, and assessed self-reported medication adherence.

For the self-management support component, interven- tion nurses delivered module-based self-management educa- tion during 16 of the 26 scheduled encounters. Each module covered a unique topic addressing knowledge and/or self-efficacy.5

For the diet/activity support component, a dietitian called participants with a body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) of 25 or greater to develop an individualized diet plan postenroll- ment. Plans were tailored to patient preferences and targeted to greater than 5% weight loss via a deficit of 500 to 750 calo- ries per day.33 Patients were also encouraged to maintain 150 minutes or more of moderate to vigorous activity weekly.34

During each of the 26 scheduled encounters, the nurse re- viewed progress. An additional dietitian phone follow-up could be arranged for patients not meeting goals.

For the medication management component, each site used 2 to 3 diabetes specialists (physicians, clinical pharma- cists, or nurse practitioners). After each of the 26 scheduled encounters, the intervention nurse forwarded an EHR-based summary note to the medication manager. The medication manager considered treatment changes with guidance from a medication protocol, which targeted a fasting glucose level of 90 to 150 mg/dL and preprandial glucose level of 140 to 180 mg/dL (tailoring permitted based on HbA1c goal/hypoglyce- mia; to convert glucose level to mmol/L, multiply by 0.0555). The medication manager conveyed recommendations via an EHR note addendum, which the nurse implemented; medi-

Figure 1. Comprehensive Telehealth Intervention Design

Scheduled phone encounter (15-30 min per encounter, every 2 wk × 26 wk)

Templated report compiled and documented in EHR

1. Telemonitoring Nurse reviews SMBG data, medications, adherence

2. Self-management support Nurse delivers self-management module

3. Diet/activity support Nurse supports individualized diet plan

and activity plan

4. Medication management Report sent to study medication manager through EHR after encounters, changes

implemented by nurse

5. Depression support Nurse screens for depression every 12 wk,

facilitates study psychiatrist assessment for positive screens

Adapted with permission from Kobe et al.5 EHR indicates electronic health record; SMBG, self-monitored blood glucose.

Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes Original Investigation Research

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine Published online July 25, 2022 E3

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

cation managers did not routinely contact participants. Pri- mary clinicians were alerted to changes via the EHR.

For the depression support component, participants with a Personal Health Questionnaire-8 (PHQ-8) score of 10 or higher at baseline or subsequent screening entered the depression pro- tocol, which was supported by 1 psychiatrist at each site and provided pharmacologic and nonpharmacologic options.35 All patients receiving depression support had PHQ-8 follow-up ev- ery 8 weeks, with treatment changes as needed.

Telemonitoring/Care Coordination Intervention Participants transmitted SMBG data and received automated self-management information daily by phone. Participants re- ceived nurse calls for alert SMBG values and could reach nurses as needed for acute issues but did not complete scheduled en- counter calls. Participants also received care coordination, in- cluding communication about upcoming appointments, notification of primary clinicians regarding acute needs, and preappointment compilation of SMBG data for review by primary clinicians. Diabetes medication management was not formally integrated into this intervention; instead, medica- tion adjustments were at the discretion of existing clinicians during or outside of scheduled encounters. Of note, because telemonitoring/care coordination is considered routine HT practice, no training was required for nurses delivering this intervention.

Outcomes Outcome assessments were performed by blinded research as- sistants at 3-month intervals for 12 months. Full assessments were competed at 0, 6, and 12 months, with additional HbA1c- only assessments at 3 and 9 months.

The primary outcome was HbA1c level. Secondary out- comes were diabetes distress,36 diabetes self-care,37

self-efficacy,38 BMI, and depression symptoms.39 Adverse events were assessed by structured self-report40; incidence of blood glucose level less than 70 mg/dL was also examined using SMBG data transmitted to HT in both arms.

Intervention Costs Intervention costs were examined in both arms. Labor costs included all intervention nurse, dietitian, medication man- ager, and psychiatrist time spent delivering the intervention; capital costs included HT equipment, telephone service costs, overhead, and supplies.

Fidelity Assessment Fidelity assessment for the comprehensive telehealth inter- vention included nurse tracking of encounters and time using online software. The principal investigator and project coor- dinator conducted periodic shadowing of nurses and bian- nual case review meetings with the medication managers.5 Up- dated medications were tracked at outcome visits.

Influence of COVID-19 Pandemic on Outcome Ascertainment In March 2020, VHA announced a pandemic-related restric- tion on in-person research interactions. Survey data collec-

tion continued by phone when possible, but study measure- ment of HbA1c level and BMI was interrupted. In May 2020, we received permission to resume collection of HbA1c level given its importance to diabetes management; however, BMI could not be measured after March 2020, which affected the 12-month time point for 169 participants. Despite the pandemic, intervention delivery continued uninterrupted at both sites.

Statistical Analyses Sample Size Per previous data,31 we used an α level of .05, 80% power, 20% dropout, within-patient correlation 0.55, SD 1.6, and baseline HbA1c level of 10.3% to estimate that 100 participants per arm would detect a clinically significant HbA1c difference of 0.6% at 12 months.41 Power estimates were derived by generating 1000 stimulated data sets with these assumptions and fitting linear mixed models to assess the effect difference at 12 months.

Analytic Approach All analyses were intention-to-treat and performed using SAS, version 9.4 (SAS Institute).42 Linear mixed longitudinal mod- els were used for all primary and secondary outcomes.43 The primary outcome model included fixed effects for linear, qua- dratic and cubic time, time-by-arm interaction terms, and ran- domization stratification variables, and random effects for in- tercept and linear time (see eMethods in Supplement 2). The covariance structure was determined using Akaike informa- tion criteria.44 Our primary inference was on the estimated be- tween-arm 12-month HbA1c difference. As a post hoc sensitiv- ity analysis, we included baseline covariates with between- arm differences in the primary model. Also, to explore a dose- response effect of the comprehensive telehealth intervention, we conducted a descriptive post hoc analysis examining HbA1c

change among participants completing more than 20 vs 20 or fewer encounters.

For secondary outcomes, fixed effects included dummy- coded time effects for each time point and time-by-arm inter- action terms. Given the pandemic’s hindrance of BMI ascer- tainment, BMI was analyzed only at 0 and 6 months. To account for within-participant repeated measures, we fit an unstruc- tured covariance. We descriptively analyzed intervention en- gagement (encounter completion, SMBG transmission), ad- verse events, and costs.

Missing Data Our analyses implicitly accommodated missingness when re- lated to prior outcome data or other baseline model covari- ates defined as missing at random (MAR). As a sensitivity analy- sis for the primary model, we also multiplied imputed missing HbA1c data using a Markov chain Monte Carlo algorithm in- corporating additional variables to strengthen the MAR as- sumption (see eMethods in Supplement 2). With the pandem- ic’s influence on outcome ascertainment, we conducted another sensitivity analysis fitting our primary model with in- clusion of additional clinical HbA1c measurements obtained from the EHR.

Research Original Investigation Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes

E4 JAMA Internal Medicine Published online July 25, 2022 (Reprinted) jamainternalmedicine.com

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

Data Monitoring Committee Our data monitoring committee comprised the study statisti- cians (A.S.J. and C.J.C.) and 3 independent experts. This com- mittee met at 6-month intervals during the study and exam- ined recruitment, retention, randomization, adverse events, and outcomes.

Results Participants, Retention, and Fidelity Participants were enrolled from December 2018 through Janu- ary 2020; participant contact concluded in January 2021. Of 1128 individuals assessed, 257 were consented, and 200 were randomized (Figure 2); most of those consented but not randomized were excluded for baseline HbA1c level of less than 8.5%. Randomized participants were similar to those declining participation (eTable 1 in Supplement 2). Of those ran- domized, 101 were allocated to comprehensive telehealth and 99 to telemonitoring/care coordination; all were analyzed in their randomized group, and there was no between-arm crossover.

Per Table 1, participants had baseline mean (SD) age of 57.8 (8.2) years, HbA1c level of 10.2% (1.3%), and BMI of 34.8 (6.7). A total of 45 (22.5%) participants were women; 144 (72.0%) were Black, 11 (5.5%) were Hispanic/Latinx, 42 (21.0%) were White, and 14 (7.0%) were of other race. Prior to enrollment, 136 (68.0%) participants had received diabetes specialty care, and 12 (6.0%) had received HT care. Baseline characteristics were generally balanced across arms; moderate between- arm imbalances were noted in race, medication use, BMI, and social support.

Overall, 150 of 200 (75%) participants completed the 12- month assessment for HbA1c level, and 137 (68.5%) for survey- based outcomes (Figure 2). Participants in the comprehen- sive telehealth arm completed an average of 19.6 of 26 possible encounters; 33 participants completed 20 or fewer encoun- ters, and 14 completed 10 or fewer. Mean (SD) encounter time was 17.0 (11.2) minutes. Telemonitoring, self-management sup- port, diet/activity support, and medication management were delivered during all completed comprehensive telehealth en- counters; 30 of 101 (29.7%) participants initiated depression support based on elevated PHQ-8 score at baseline, and 23 of 90 (25.6%) at 6 months. Because the telemonitoring/care co- ordination intervention did not involve scheduled encoun- ters, encounter metrics were not tracked. Both groups expe- rienced changes in medication use during the study (eTable 2 in Supplement 2); descriptively, the comprehensive tele- health group had greater increases in use of glucagon-like pep- tide-1 receptor agonists (+15% vs +8%) and dipeptidyl pepti- dase-4 inhibitors (+5% vs +1%) from baseline to 12 months.

Primary Outcome Between baseline and 12 months, estimated HbA1c change was −1.59% (10.17% to 8.58%) in the comprehensive telehealth group and −0.98% (10.17% to 9.19%) in the telemonitoring/ care coordination group; the estimated difference of −0.61% (95% CI, −1.12% to −0.11%; P = .02) favored comprehensive tele-

health (Table 2, Figure 3). Three-way interactions between in- tervention arm, time, and key stratification variables were not statistically significant, indicating no evidence for differen- tial HbA1c effects over time based on preenrollment diabetes specialty care or study site.

Similar results were found on sensitivity analyses with MAR imputation using multiply imputed data sets (mean 12-month difference, −0.63%; 95% CI, −0.95% to −0.35%; P = .03) and inclusion of additional clinical HbA1c measures from the study period (n = 191 from 62 comprehensive tele- health and 63 telemonitoring/care coordination participants; mean 12-month difference, −0.50%; 95% CI, −0.99% to −0.01%; P = .04). Findings with baseline covariate adjust- ment (race; insulin, metformin, sulfonylurea, sodium- glucose cotransporter-2 inhibitor use; BMI; social support) were also similar (mean 12-month difference, −0.66%; 95% CI, −1.17% to −0.14%; P = .01).

On exploratory descriptive analyses (eTable 3 in Supple- ment 2), comprehensive telehealth patients who completed

Figure 2. Participant Flow

1128 Patients assessed for eligibility

257 Consented

871 Excluded 458 Ineligible for the study

1 Eligible, but goal sample reached

293 Refused to participate 119 Unable to contact

200 Randomized

101 Randomized to comprehensive telehealth intervention

52 Excluded from study 48 HbA1c <8.5%

1 Did not receive VHA care

1 Upcoming bariatric surgery 1 Hospitalized

1 Invalid HbA1c from laboratory 4 Unable to contact 1 Withdrew from the study

HbA1c

82 6 mo

101 Baseline 87 3 mo

BMI

78 9 mo 77 12 mo

15 12 mo

101 Baseline 67 6 mo

89 6 mo

Survey measures 101 Baseline

71 12 mo

99 Randomized to telemonitoring/ care coordination

HbA1c

87 6 mo

99 Baseline 85 3 mo

BMI

75 9 mo 73 12 mo

16 12 mo

99 Baseline 68 6 mo

91 6 mo

Survey measures 99 Baseline

66 12 mo

BMI indicates body mass index, calculated as weight in kilograms divided by height in meters squared; HbA1c, hemoglobin A1c; VHA, Veterans Health Administration.

Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes Original Investigation Research

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine Published online July 25, 2022 E5

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

more than 20 encounters (n = 68) experienced greater HbA1c

reduction (1.84%) than those who completed 20 or fewer (n = 33, 0.79%).

Secondary Outcomes Per Table 2, the comprehensive telehealth group had a greater improvement than the telemonitoring/care coordination group at 12 months for diabetes distress (mean difference, −0.25; 95% CI, −0.42 to −0.07), diabetes self-care (mean difference, 0.51; 95% CI, 0.25 to 0.78), and self-efficacy (mean difference, 0.39; 95% CI, 0.07 to 0.71). There were no statistically significant be- tween-group differences in depressive symptoms (12 months) or BMI (6 months).

Adverse Events Adverse events were similar between arms. Comprehensive telehealth participants had 26 serious events (16 hospital- izations), 3 possibly study-related (episodes of ketoacidosis, hyperglycemia, and possible medication-related urinary infection). In the telemonitoring/care coordination group, there were 19 serious events (17 hospitalizations, 1 death), none deemed study-related. Among the 97 comprehensive telehealth participants with available SMBG data, 70 (72.1%) reported 1 or more SMBG values less than 70 mg/dL over 12 months, with a per-patient mean (SD) of 7.3 (9.8). Among the 89 telemonitoring/care coordination participants with available SMBG data, 61 (68.5%) reported 1 or more SMBG

Table 1. Baseline Characteristics, Overall and Stratified by Study Arm

Baseline characteristics

No. (%)

Overall (n = 200) Comprehensive telehealth (n = 101)

Telemonitoring/care coordination (n = 99)

Demographic characteristics

Age, mean (SD), y 57.8 (8.2) 57.7 (8.3) 57.8 (8.0)

Sex

Female 45 (22.5) 24 (23.8) 21 (21.2)

Male 155 (77.5) 77 (76.2) 78 (78.8)

Race

Black or African American 144 (72.0) 68 (67.3) 76 (76.8)

White 42 (21.0) 25 (24.8) 17 (17.2)

Other racea 14 (7.0) 8 (7.9) 6 (6.0)

Hispanic/Latino ethnicityb 11 (5.5) 6 (5.9) 5 (5.1)

Did not graduate high school 57 (28.5) 29 (28.7) 28 (28.3)

Currently married 91 (45.5) 45 (46.5) 46 (45.5)

Social supportc,d 191 (95.5) 98 (97.0) 93 (93.9)

Employed (full/part-time/self) 90 (45.0) 48 (47.5) 42 (42.4)

Study site

Durham 115 (57.5) 57 (56.4) 58 (58.6)

Richmond 85 (42.5) 44 (43.6) 41 (41.4)

Years with diabetes, mean (SD) 12.1 (7.7) 12.1 (8.0) 12.0 (7.5)

Prior diabetes specialty care 136 (68.0) 69 (68.3) 67 (67.7)

Prior Home Telehealth enrollment 12 (6.0) 6 (5.9) 6 (6.1)

Clinical measures

Baseline HbA1c, mean (SD), % 10.2 (1.3) 10.1 (1.2) 10.2 (1.4)

BMI, mean (SD) 34.8 (6.7) 34.5 (6.4) 35.2 (7.0)

Hypertension 166 (83.0) 81 (80.2) 85 (85.9)

Hyperlipidemiad 171 (85.5) 84 (83.2) 87 (87.9)

Tobacco use in past 6 mo 30 (25.4) 16 (30.2) 14 (21.5)

Metformin 160 (80.0) 78 (77.2) 82 (82.8)

Sulfonylurea 83 (41.5) 35 (34.7) 48 (48.9)

Thiazolidinedione 14 (7.0) 7 (6.9) 7 (7.1)

SGLT-2 inhibitor 22 (11.0) 15 (14.9) 7 (7.1)

GLP-1 receptor agonist 25 (12.5) 11 (10.9) 14 (14.1)

DPP-4 inhibitor 4 (2.0) 0 4 (4.0)

Insulin use 142 (71.0) 78 (77.2) 64 (64.6)

Psychosocial measures

Diabetes distress (DDS), mean (SD) 1.9 (0.8) 1.9 (0.7) 1.9 (0.9)

Diabetes self-care (DSMQ), mean (SD) 6.7 (1.6) 6.9 (1.5) 6.5 (1.7)

Self-efficacy (PCS), mean (SD) 5.2 (1.5) 5.2 (1.5) 5.2 (1.4)

Depression (PHQ-8) score, mean (SD)e 7.3 (5.7) 7.0 (5.2) 7.6 (6.1)

Abbreviations: BMI, body mass index, calculated as weight in kilograms divided by height in meters squared; DDS, Diabetes Distress Scale; DPP-4, dipeptidyl peptidase-4; DSMQ, Diabetes Self-Management Questionnaire; GLP-1, glucagon-like peptide-1; HbA1c, hemoglobin A1c; PCS, Perceived Competence Scale; PHQ-8, Patient Health Questionnaire-8; SGLT-2, sodium-glucose cotransporter-2. a Because of low numbers in the

Asian, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, other, and unknown categories, these were combined into a single category, “other race.”

b One patient in the telemonitoring/care coordination group responded “Don’t know” to the Hispanic/Latino ethnicity question.

c Social support was assessed by asking, “Do you have someone you feel close to, someone you can trust and confide in?”

d One patient in the comprehensive telehealth group responded “Don’t know” to having high cholesterol and to having social support.

e One patient in the comprehensive telehealth group and 1 patient in the telemonitoring/care coordination group were missing the PHQ-8 score.

Research Original Investigation Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes

E6 JAMA Internal Medicine Published online July 25, 2022 (Reprinted) jamainternalmedicine.com

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

values less than 70 mg/dL over 12 months, with a per- patient mean (SD) of 6.9 (11.1).

Intervention Costs Per-patient intervention costs were $2465 for comprehen- sive telehealth and $946 for telemonitoring/care coordina- tion, for a between-arm difference of $1519 over 12 months.

Discussion We sought to promote practical use of telehealth for PPDM by comparing 2 approaches designed for feasible clinical imple- mentation: a comprehensive telehealth intervention and telemonitoring/care coordination. While both approaches improved HbA1c level, the comprehensive telehealth inter- vention produced a greater 12-month improvement in HbA1c

level and multiple secondary outcomes, without excess hypoglycemia.

These findings demonstrate that practically designed tele- health can be effective for patients whose T2D remains per- sistently poorly controlled despite clinic-based care. More- over, combining telehealth strategies to target multiple barriers to improvement lowers HbA1c level more than simpler ap- proaches like telemonitoring/care coordination. Given our ac-

tive-comparator design, we cannot exclude that the within- arm HbA1c effects (−1.59% for comprehensive telehealth and −0.98% for telemonitoring/care coordination) may partly re- flect regression to the mean; for example, the −0.98% HbA1c

improvement with telemonitoring/care coordination ex- ceeds the effect reported for telemonitoring in systematic re- views (−0.4% to 0.5% vs usual care).15,45-47 However, the rela- tive HbA1c benefit seen with comprehensive telehealth in this study (−0.61%) was not subject to regression to the mean. Im- portantly, the HbA1c reduction with comprehensive tele- health was durably retained through 12 months.

While the comprehensive telehealth intervention was more expensive (additional $1519 over 12 months), this incremen- tal cost is less than most branded glucose-lowering medica- tions, and the comprehensive approach came with added ben- efits for diabetes distress, self-care, and self-efficacy. Given the high complication rates characteristic of PPDM and the long- term cost benefits of HbA1c reduction,1,2 implementing com- prehensive telehealth in practice may represent an appropri- ate investment for health care systems in which the requisite infrastructure is or can be made available.

When we developed this study, our focus was on leverag- ing the VHA’s telehealth infrastructure to generate solutions for PPDM. Since then, the COVID-19 pandemic has driven a dra- matic upsurge in telehealth use worldwide. Currently,

Table 2. Estimated Outcome Means and Mean Differences for Comprehensive Telehealth (n = 101) and Telemonitoring/Care Coordination (n = 99) Arms by Time Pointa

Outcome

Estimated mean Estimated mean difference (95% CI) P value

Comprehensive telehealth

Telemonitoring/care coordination

HbA1c, %

Baseline 10.17 10.17 NA NA

3 mo 8.90 9.18 −0.29 (−0.48 to −0.09) NA

6 mo 8.54 9.03 −0.48 (−0.78 to −0.18) NA

9 mo 8.61 9.20 −0.60 (−0.96 to −0.22) NA

12 mo 8.58 9.19 −0.61 (−1.12 to −0.11) .02

Diabetes distress (DDS)b

Baseline 1.93 1.93 NA NA

6 mo 1.53 1.57 −0.04 (−0.18 to 0.09) NA

12 mo 1.43 1.67 −0.25 (−0.42 to −0.07) .007

Diabetes self-care (DSMQ)c

Baseline 6.67 6.67 NA NA

6 mo 8.15 7.92 0.22 (−0.07 to 0.51) NA

12 mo 8.34 7.83 0.51 (0.25 to 0.78) <.001

Self-efficacy (PCS)d

Baseline 5.20 5.20 NA NA

6 mo 6.09 5.84 0.24 (−0.06 to 0.54) NA

12 mo 6.31 5.92 0.39 (0.07 to 0.71) .02

BMI

Baseline 34.81 34.81 NA NA

6 mo 35.05 34.86 0.19 (−0.24 to 0.62) .39

Depression symptoms (PHQ-8)e

Baseline 7.32 7.32 NA NA

6 mo 6.54 6.06 0.48 (−0.72 to 1.69) NA

12 mo 4.64 5.80 −1.16 (−2.53 to 0.21) .10

Abbreviations: BMI, body mass index, calculated as weight in kilograms divided by height in meters squared; DDS, Diabetes Distress Scale; DSMQ, Diabetes Self-Management Questionnaire; HbA1c, hemoglobin A1c; NA, not applicable; PCS, Perceived Competence Scale; PHQ-8, Patient Health Questionnaire-8. a Missing data by time point for the

comprehensive telehealth group were as follows: HbA1c: 3 months n = 14, 6 months n = 19, 9 months n = 23, 12 months n = 22; survey measures: 6 months n = 12, 12 months n = 30; BMI: 6 months n = 34. Missing data by time point for the telemonitoring/care coordination group were as follows: HbA1c: 3 months n = 14, 6 months n = 12, 9 months n = 24, 12 months n = 26; survey measures: 6 months n = 10, 12 months n = 35; BMI: 6 months n = 33. No data points were missing at baseline.

b A lower score on the DDS indicates lower levels of diabetes distress, so is preferred.

c A higher score on the DSMQ indicates better diabetes self-care, so is preferred.

d A higher score on the PCS indicates higher self-efficacy, so is preferred.

e A lower score on the PHQ-8 indicates fewer depressive symptoms, so is preferred.

Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes Original Investigation Research

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine Published online July 25, 2022 E7

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

telehealth, as used by most systems, implies intermittent video or phone appointments,30,48-50 with limited interim patient– clinician contact or transfer of patient-generated data into the EHR—in essence, “clinic-like” care, only delivered remotely. Just as clinic-based care does not fully address the factors un- derlying PPDM, “clinic-like” telehealth is also likely inad- equate. There has always been a sound argument for using comprehensive telehealth when clinic-based chronic disease care falls short; now that telehealth has gained wider accep- tance, systems have a clear mandate to maximize its value for those high-risk patients who respond insufficiently to clinic care. The present study provides evidence supporting com- prehensive telehealth for PPDM within the VHA, but also pre- sents a template for how other systems might use existing re- sources to improve the management of PPDM and other hard- to-treat conditions.

While our focus on examining practical, comprehensive telehealth specifically for PPDM is novel, the findings also add to the broader telehealth literature in T2D. The between-arm HbA1c effect we observed (−0.61%) is notable given our active- comparator design and the relatively modest effects reported in recent systematic reviews of telehealth interventions (mean HbA1c benefit vs usual care, −0.4%).51,52 In particular, prior RCTs examining comprehensive interventions for T2D have not low- ered HbA1c level20-25; these include studies that specifically

sought to examine multicomponent interventions vs usual care in pragmatic settings (non-VHA), with neutral results.20-22

Limitations Despite efforts to oversample women, the population demo- graphics may limit generalizability. However, the cohort’s high proportion of Black participants (72.0%) is a strength and may suggest that the pandemic-induced shift to telehealth need not exacerbate health care inequities.53,54 The studied interven- tions were designed for practical delivery within the VHA, which may limit generalizability to systems lacking capacity for nurse-delivered telehealth, integrated mental health, and dietitian services; nevertheless, the idea of designing tele- health interventions to leverage available resources is broadly applicable. The VHA costs may not fully generalize, but the be- tween-arm cost difference may be more translatable.

While clinical intervention delivery continued unim- peded during the pandemic, the missing data frequency was higher than expected at 12 months; however, the sensitivity analyses with multiple imputation and inclusion of clinical HbA1c data support the validity of the findings. Of note, par- ticipants could not be blinded to randomization status, which leaves potential for bias, especially pertaining to subjective sur- vey measures.

This study was not designed to evaluate the effective- ness of each individual component of the comprehensive in- tervention. Future mediator analyses will examine how each component contributes to the overall intervention effect. Fi- nally, the comprehensive telehealth intervention does not ac- count for all contributors to PPDM, including social determi- nants of health.

Conclusions Findings from this randomized clinical trial showed that com- pared with a simpler telehealth approach, a comprehensive telehealth intervention improved HbA1c level and other out- comes in patients with PPDM. Because this comprehensive telehealth intervention was delivered by clinical staff using ex- isting resources, it may warrant clinical implementation in sys- tems with appropriate infrastructure. More broadly, this study provides valuable comparative evidence that may help sys- tems maximize the value of telehealth during the COVID-19 pandemic and beyond.

ARTICLE INFORMATION

Accepted for Publication: May 31, 2022.

Published Online: July 25, 2022. doi:10.1001/jamainternmed.2022.2947

Author Affiliations: Durham Veterans Affairs Center of Innovation to Accelerate Discovery and Practice Transformation (ADAPT), Durham, North Carolina (Crowley, Bosworth, Jeffreys, Coffman, Maciejewski, Steinhauser, Smith, Strawbridge, Majette Elliott, Danus, Edelman); Division of Endocrinology, Department of Medicine, Duke University School of Medicine, Durham, North Carolina (Crowley); Central Virginia Veterans Affairs

Health Care System, Richmond (Tarkington, Fredrickson, Mundy, Marcano, Overby); Division of General Internal Medicine, Department of Medicine, Duke University School of Medicine, Durham, North Carolina (Bosworth, Maciejewski, Smith, Edelman); Department of Population Health Sciences, Duke University School of Medicine, Durham, North Carolina (Bosworth, Maciejewski, Steinhauser, Smith); Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina (Coffman); Duke-Margolis Center for Health Policy, Duke University School of Medicine, Durham, North Carolina (Maciejewski); Greenville VA Health Care

Center, Greenville, North Carolina (Dar); Division of Endocrinology, Department of Medicine, Brody School of Medicine at East Carolina University, Greenville, North Carolina (Dar).

Author Contributions: Drs Crowley and Coffman had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Crowley, Tarkington, Bosworth, Steinhauser, Smith, Strawbridge, Majette Elliott, Edelman. Acquisition, analysis, or interpretation of data: Crowley, Tarkington, Jeffreys, Coffman, Maciejewski, Steinhauser, Smith, Dar, Fredrickson,

Figure 3. Estimated Trajectories by Arm for Hemoglobin A1c (HbA1c) Level, From Linear Mixed Models

11

10

9

8

Es tim

at ed

H bA

1c , %

Time, mo 0 3 6 9 12

Comprehensive telehealth

Telemonitoring/care coordination

Estimated difference at 12 mo (comprehensive telehealth – telemonitoring/care coordination): –0.61%; 95% CI, –1.12% to –0.11%

Error bars indicate 95% CIs.

Research Original Investigation Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes

E8 JAMA Internal Medicine Published online July 25, 2022 (Reprinted) jamainternalmedicine.com

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

Mundy, Marcano, Overby, Danus. Drafting of the manuscript: Crowley, Tarkington, Bosworth, Coffman, Steinhauser, Strawbridge. Critical revision of the manuscript for important intellectual content: Tarkington, Bosworth, Jeffreys, Coffman, Maciejewski, Steinhauser, Smith, Dar, Fredrickson, Mundy, Marcano, Overby, Majette Elliott, Danus, Edelman. Statistical analysis: Jeffreys, Coffman, Maciejewski, Smith. Obtained funding: Crowley, Smith. Administrative, technical, or material support: Crowley, Tarkington, Jeffreys, Maciejewski, Mundy, Strawbridge, Marcano, Majette Elliott, Danus. Supervision: Crowley, Tarkington, Bosworth, Smith, Edelman.

Conflict of Interest Disclosures: Dr Crowley reported grants from National Institutes of Health (1R01NR019594-01), VA Quality Enhancement Research Initiative (QUE 20-012), VA Office of Rural Health, and VA Health Services Research & Development (CDA 13-261) outside the submitted work. Dr Bosworth reported research support from Otsuka, Novo Nordisk, Sanofi, Improved Patient Outcomes, Boehringer Ingelheim, National Institutes of Health, and VA, as well as consulting fees from WebMD, Sanofi, Novartis, Otsuka, Abbott, Xcenda, Preventric Diagnostics, VIDYA, and the Medicines Company outside the submitted work. Ms Danus reported grants from Department of Veterans Affairs/VA Health Services Research and Novo Nordisk outside the submitted work. Dr Edelman reported personal fees from Department of Veterans Affairs (salary) outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by a grant from Veterans Affairs Health Services Research and Development (VA IIR 16-213, Crowley PI). Dr Crowley acknowledges funding from the National Institutes of Health (1R01NR019594-01), the Veterans Affairs Quality Enhancement Research Initiative (VA QUE 20-012), and the Veterans Affairs Office of Rural Health; he was supported by a Career Development Award from Veterans Affairs Health Services Research and Development (CDA 13-261) during part of the study period. Drs Bosworth and Maciejewski were supported by Veterans Affairs Health Services Research and Development Senior Career Scientist Awards (VA HSR&D 08-027, VA HSR&D 10-391). The authors acknowledge in-kind support from the Durham Center of Innovation to Accelerate Discovery and Practice Transformation (VA CIN 13-410) within the Durham VA Health Care System.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of Veterans Affairs.

Meeting Presentation: Portions of the data reported in this manuscript were previously reported at the 81st American Diabetes Association Scientific Sessions; June 25-29, 2021; virtual.

Data Sharing Statement: See Supplement 3.

Additional Contributions: The authors would like to thank the following additional individuals for their contributions to this study: Deborah H. Jeter, RN (coordinated activities at Richmond, Virginia site); Steven Szabo, MD, PhD, and Shivan Desai, MD (study psychiatrists); Mary P. Garrett, MSN, RN, and Theresa C. Wilmot, RN (study intervention nurses); Melissa Durkee, PharmD, Susan Bullard, PharmD, and Janette Hiner, NP (study medication managers); and Teresa Howard, AA (research assistant at Durham, North Carolina site). Additional contributors were not directly compensated, but study funding (VA IIR 16-213, Crowley PI) provided partial salary support for Mss Jeter, Garrett, Wilmot, and Howard.

REFERENCES

1. Gilmer TP, O’Connor PJ, Rush WA, et al. Predictors of health care costs in adults with diabetes. Diabetes Care. 2005;28(1):59-64. doi:10. 2337/diacare.28.1.59

2. McBrien KA, Manns BJ, Chui B, et al. Health care costs in people with diabetes and their association with glycemic control and kidney function. Diabetes Care. 2013;36(5):1172-1180. doi:10.2337/dc12-0862

3. Stratton IM, Adler AI, Neil HA, et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ. 2000;321 (7258):405-412. doi:10.1136/bmj.321.7258.405

4. Alexopoulos AS, Jackson GL, Edelman D, et al. Clinical factors associated with persistently poor diabetes control in the Veterans Health Administration: a nationwide cohort study. PLoS One. 2019;14(3):e0214679. doi:10.1371/journal.pone. 0214679

5. Kobe EA, Edelman D, Tarkington PE, et al. Practical telehealth to improve control and engagement for patients with clinic-refractory diabetes mellitus (PRACTICE-DM): protocol and baseline data for a randomized trial. Contemp Clin Trials. 2020;98:106157. doi:10.1016/j.cct.2020.106157

6. Davidson J. Strategies for improving glycemic control: effective use of glucose monitoring. Am J Med. 2005;118(suppl 9A):27S-32S. doi:10.1016/j. amjmed.2005.07.054

7. Donnelly LA, Morris AD, Evans JM; DARTS/MEMO collaboration. Adherence to insulin and its association with glycaemic control in patients with type 2 diabetes. QJM. 2007;100(6): 345-350. doi:10.1093/qjmed/hcm031

8. Hill-Briggs F, Gary TL, Bone LR, Hill MN, Levine DM, Brancati FL. Medication adherence and diabetes control in urban African Americans with type 2 diabetes. Health Psychol. 2005;24(4):349- 357. doi:10.1037/0278-6133.24.4.349

9. American Diabetes Association. 7. Obesity management for the treatment of type 2 diabetes. Diabetes Care. 2017;40(suppl 1):S57-S63. doi:10. 2337/dc17-S010

10. Chiu CJ, Wray LA. Factors predicting glycemic control in middle-aged and older adults with type 2 diabetes. Prev Chronic Dis. 2010;7(1):A08.

11. Hartz A, Kent S, James P, Xu Y, Kelly M, Daly J. Factors that influence improvement for patients with poorly controlled type 2 diabetes. Diabetes Res Clin Pract. 2006;74(3):227-232. doi:10.1016/j. diabres.2006.03.023

12. de Groot M, Anderson R, Freedland KE, Clouse RE, Lustman PJ. Association of depression and diabetes complications: a meta-analysis. Psychosom Med. 2001;63(4):619-630. doi:10.1097/ 00006842-200107000-00015

13. Morrison F, Shubina M, Turchin A. Encounter frequency and serum glucose level, blood pressure, and cholesterol level control in patients with diabetes mellitus. Arch Intern Med. 2011;171(17): 1542-1550. doi:10.1001/archinternmed.2011.400

14. Rosenthal ES, Bashan E, Herman WH, Hodish I. The effort required to achieve and maintain optimal glycemic control. J Diabetes Complications. 2011;25 (5):283-288. doi:10.1016/j.jdiacomp.2011.01.002

15. Medical Advisory Secretariat. Home telemonitoring for type 2 diabetes: an evidence-based analysis. Ont Health Technol Assess Ser. 2009;9(24):1-38.

16. Medical Advisory Secretariat. Behavioural interventions for type 2 diabetes: an evidence-based analysis. Ont Health Technol Assess Ser. 2009;9(21):1-45.

17. Kempf K, Altpeter B, Berger J, et al. Efficacy of the telemedical lifestyle intervention program TeLiPro in advanced stages of type 2 diabetes: a randomized controlled trial. Diabetes Care. 2017; 40(7):863-871. doi:10.2337/dc17-0303

18. Pimouguet C, Le Goff M, Thiébaut R, Dartigues JF, Helmer C. Effectiveness of disease-management programs for improving diabetes care: a meta-analysis. CMAJ. 2011;183(2):E115-E127. doi: 10.1503/cmaj.091786

19. Atlantis E, Fahey P, Foster J. Collaborative care for comorbid depression and diabetes: a systematic review and meta-analysis. BMJ Open. 2014;4(4): e004706. doi:10.1136/bmjopen-2013-004706

20. Lauffenburger JC, Ghazinouri R, Jan S, et al. Impact of a novel pharmacist-delivered behavioral intervention for patients with poorly-controlled diabetes: the ENhancing outcomes through Goal Assessment and Generating Engagement in Diabetes Mellitus (ENGAGE-DM) pragmatic randomized trial. PLoS One. 2019;14(4):e0214754. doi:10.1371/journal.pone.0214754

21. Sarayani A, Mashayekhi M, Nosrati M, et al. Efficacy of a telephone-based intervention among patients with type-2 diabetes: a randomized controlled trial in pharmacy practice. Int J Clin Pharm. 2018;40(2):345-353. doi:10.1007/s11096-018- 0593-0

22. Edelman D, Dolor RJ, Coffman CJ, et al. Nurse-led behavioral management of diabetes and hypertension in community practices: a randomized trial. J Gen Intern Med. 2015;30(5): 626-633. doi:10.1007/s11606-014-3154-9

23. Lee JY, Chan CKY, Chua SS, et al. Telemonitoring and team-based management of glycemic control on people with type 2 diabetes: a cluster-randomized controlled trial. J Gen Intern Med. 2020;35(1):87-94. doi:10.1007/s11606-019- 05316-9

24. Crowley MJ, Powers BJ, Olsen MK, et al. The Cholesterol, Hypertension, And Glucose Education (CHANGE) study: results from a randomized controlled trial in African Americans with diabetes. Am Heart J. 2013;166(1):179-186. doi:10.1016/j.ahj. 2013.04.004

25. Tang PC, Overhage JM, Chan AS, et al. Online disease management of diabetes: engaging and

Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes Original Investigation Research

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine Published online July 25, 2022 E9

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

motivating patients online with enhanced resources-diabetes (EMPOWER-D), a randomized controlled trial. J Am Med Inform Assoc. 2013;20(3): 526-534. doi:10.1136/amiajnl-2012-001263

26. Adler-Milstein J, Kvedar J, Bates DW. Telehealth among US hospitals: several factors, including state reimbursement and licensure policies, influence adoption. Health Aff (Millwood). 2014;33(2):207-215. doi:10.1377/hlthaff.2013.1054

27. Glasgow RE, Lichtenstein E, Marcus AC. Why don’t we see more translation of health promotion research to practice? rethinking the efficacy-to-effectiveness transition. Am J Public Health. 2003;93(8):1261-1267. doi:10.2105/AJPH.93. 8.1261

28. Scott Kruse C, Karem P, Shifflett K, Vegi L, Ravi K, Brooks M. Evaluating barriers to adopting telemedicine worldwide: a systematic review. J Telemed Telecare. 2018;24(1):4-12. doi:10.1177/ 1357633X16674087

29. Ross J, Stevenson F, Lau R, Murray E. Factors that influence the implementation of e-health: a systematic review of systematic reviews (an update). Implement Sci. 2016;11(1):146. doi:10.1186/ s13012-016-0510-7

30. Wosik J, Fudim M, Cameron B, et al. Telehealth transformation: COVID-19 and the rise of virtual care. J Am Med Inform Assoc. 2020;27(6):957-962. doi:10.1093/jamia/ocaa067

31. Crowley MJ, Edelman D, McAndrew AT, et al. Practical telemedicine for veterans with persistently poor diabetes control: a randomized pilot trial. Telemed J E Health. 2016;22(5):376-384. doi:10.1089/tmj.2015.0145

32. Draznin B, Aroda VR, Bakris G, et al; American Diabetes Association Professional Practice Committee. 6. Glycemic targets: Standards of Medical Care in Diabetes—2022. Diabetes Care. 2022;45(S1)(suppl 1):S83-S96.

33. Draznin B, Aroda VR, Bakris G, et al; American Diabetes Association Professional Practice Committee. 8. Obesity and weight management for the prevention and treatment of type 2 diabetes: Standards of Medical Care in Diabetes—2022. Diabetes Care. 2022;45(S1)(suppl 1):S113-S124.

34. Draznin B, Aroda VR, Bakris G, et al; American Diabetes Association Professional Practice Committee; American Diabetes Association Professional Practice Committee. 5. Facilitating behavior change and well-being to improve health outcomes: Standards of Medical Care in Diabetes—2022. Diabetes Care. 2022;45(S1)(suppl 1):S60-S82.

35. US Department of Veterans Affairs. VA/DoD clinical practice guideline for the management of major depressive disorder. Accessed May 4, 2022. https://www.healthquality.va.gov/guidelines/MH/ mdd/VADoDMDDCPGFINAL82916.pdf

36. Polonsky WH, Fisher L, Earles J, et al. Assessing psychosocial distress in diabetes: development of the diabetes distress scale. Diabetes Care. 2005;28 (3):626-631. doi:10.2337/diacare.28.3.626

37. Schmitt A, Gahr A, Hermanns N, Kulzer B, Huber J, Haak T. The Diabetes Self-Management Questionnaire (DSMQ): development and evaluation of an instrument to assess diabetes self-care activities associated with glycaemic control. Health Qual Life Outcomes. 2013;11:138. doi:10.1186/1477-7525-11-138

38. Williams GC, Freedman ZR, Deci EL. Supporting autonomy to motivate patients with diabetes for glucose control. Diabetes Care. 1998;21(10):1644- 1651. doi:10.2337/diacare.21.10.1644

39. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The PHQ-8 as a measure of current depression in the general population. J Affect Disord. 2009;114(1-3):163-173. doi:10.1016/j. jad.2008.06.026

40. Bent S, Padula A, Avins AL. Brief communication: Better ways to question patients about adverse medical events: a randomized, controlled trial. Ann Intern Med. 2006;144(4):257- 261. doi:10.7326/0003-4819-144-4-200602210- 00007

41. Sacks DB, Arnold M, Bakris GL, et al; National Academy of Clinical Biochemistry; Evidence-Based Laboratory Medicine Committee of the American Association for Clinical Chemistry. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Diabetes Care. 2011;34(6):e61-e99. doi:10.2337/ dc11-9998

42. Little R, Kang S. Intention-to-treat analysis with treatment discontinuation and missing data in clinical trials. Stat Med. 2015;34(16):2381-2390. doi: 10.1002/sim.6352

43. Fitzmaurice GM, Laird NM, Ware JH. Applied Longitudinal Analysis. 2nd ed. Wiley; 2011. doi:10. 1002/9781119513469

44. Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr. 1974;19 (6):716-723. doi:10.1109/TAC.1974.1100705

45. Zhu X, Williams M, Finuf K, et al. Home telemonitoring of patients with type 2 diabetes:

a meta-analysis and systematic review. Diabetes Spectr. 2022;35(1):118-128. doi:10.2337/ds21-0023

46. Kim Y, Park JE, Lee BW, Jung CH, Park DA. Comparative effectiveness of telemonitoring versus usual care for type 2 diabetes: a systematic review and meta-analysis. J Telemed Telecare. 2019;25(10): 587-601. doi:10.1177/1357633X18782599

47. Lee PA, Greenfield G, Pappas Y. The impact of telehealth remote patient monitoring on glycemic control in type 2 diabetes: a systematic review and meta-analysis of systematic reviews of randomised controlled trials. BMC Health Serv Res. 2018;18(1):495. doi:10.1186/s12913-018-3274-8

48. Baum A, Kaboli PJ, Schwartz MD. Reduced in-person and increased telehealth outpatient visits during the COVID-19 pandemic. Ann Intern Med. 2021;174(1):129-131. doi:10.7326/M20-3026

49. Heyworth L, Kirsh S, Zulman D, Ferguson JM, Kizer KW. Expanding access through virtual care: the VA’s early experience with COVID-19. NEJM Catalyst. July 1, 2020. Accessed January 18, 2022. https://catalyst.nejm.org/doi/full/10.1056/cat.20. 0327

50. Artandi M, Thomas S, Shah NR, Srinivasan M. Rapid system transformation to more than 75% primary care video visits within three weeks at Stanford: response to public safety crisis during a pandemic. NEJM Catalyst. April 21, 2020. Accessed January 18, 2022. https://catalyst.nejm.org/doi/full/ 10.1056/CAT.20.0100

51. Hangaard S, Laursen SH, Andersen JD, et al. The effectiveness of telemedicine solutions for the management of type 2 diabetes: a systematic review, meta-analysis, and meta-regression. J Diabetes Sci Technol. Published online December 26, 2021. doi:10.1177/19322968211064633

52. Faruque LI, Wiebe N, Ehteshami-Afshar A, et al; Alberta Kidney Disease Network. Effect of telemedicine on glycated hemoglobin in diabetes: a systematic review and meta-analysis of randomized trials. CMAJ. 2017;189(9):E341-E364. doi:10.1503/cmaj.150885

53. Litchfield I, Shukla D, Greenfield S. Impact of COVID-19 on the digital divide: a rapid review. BMJ Open. 2021;11(10):e053440. doi:10.1136/bmjopen- 2021-053440

54. Eruchalu CN, Pichardo MS, Bharadwaj M, et al. The expanding digital divide: digital health access inequities during the COVID-19 pandemic in New York City. J Urban Health. 2021;98(2):183-186. doi: 10.1007/s11524-020-00508-9

Research Original Investigation Comprehensive Telehealth Intervention vs Telemonitoring and Care Coordination in Type 2 Diabetes

E10 JAMA Internal Medicine Published online July 25, 2022 (Reprinted) jamainternalmedicine.com

© 2022 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ by a Duke Medical Center Library User on 07/25/2022

Original Paper

Efficacy of Telemedicine in Hypertension Care Through Home Blood Pressure Monitoring and Videoconferencing: Randomized Controlled Trial

Junichi Yatabe1, MD, PhD; Midori Sasaki Yatabe1, MD, PhD; Rika Okada2, MD; Atsuhiro Ichihara3, MD, PhD 1General Incorporated Association TelemedEASE, Shinjuku-ku, Tokyo, Japan 2TM Clinic Nishishinjuku, Shinjuku-ku, Tokyo, Japan 3Department of Endocrinology and Hypertension, Tokyo Women’s Medical University, Shinjuku-ku, Tokyo, Japan

Corresponding Author: Junichi Yatabe, MD, PhD General Incorporated Association TelemedEASE 201 FARO Kagurazaka, 5-1 Fukuromachi Shinjuku-ku, Tokyo, 162-0828 Japan Phone: 81 3 6869 6938 Fax: 81 3 6478 8301 Email: [email protected]

Abstract

Background: The burden of time is often the primary reason why patients discontinue their treatment. Telemedicine may help patients adhere to treatment by offering convenience.

Objective: This study examined the efficacy and safety of telemedicine for the management of hypertension in Japan.

Methods: Patients with uncomplicated hypertension were recruited through web advertising between November 2015 and February 2017. They were then screened, stratified by office systolic blood pressure (SBP), and randomized into two groups: usual care (UC) and telemedicine. The telemedicine group used a 3G network–attached home blood pressure (BP) monitoring device, consulted hypertension specialists from an academic hospital through web-based video visits, and received prescription medication by mail for 1 year. The UC group used the same BP monitoring device but was managed using self-recorded BP readings, which included their diary entries and office BP taken in a community practice setting.

Results: Initial screening was completed by 99 patients, 54% of whom had untreated hypertension. Baseline BP was similar between the groups, but the weekly average SBP at the end of the 1-year study period was significantly lower in the telemedicine group (125, SD 9 mmHg vs 131, SD 12 mmHg, respectively; P=.02). SBP in the telemedicine group was 3.4 mmHg lower in the morning and 5.8 mmHg lower in the evening. The rate of SBP control (135 mmHg) was better in the telemedicine group (85.3% vs 70.0%; P=.01), and significant adverse events were not observed.

Conclusions: We present evidence suggesting that antihypertensive therapy via home BP telemonitoring and web-based video visits achieve better BP control than conventional care and is a safe treatment alternative that warrants further investigation.

Trial Registration: UMIN-CTR UMIN000025372; https://tinyurl.com/47ejkn4b

(JMIR Cardio 2021;5(2):e27347) doi: 10.2196/27347

KEYWORDS

blood pressure management; digital health; web-based medicine; prospective study; telemonitoring; blood pressure; monitoring; telemedicine; telehealth; efficacy; hypertension; video conference; safety; Japan

Introduction

Antihypertensive therapy has advanced over the years to enable lowering of blood pressure (BP) in most patients with hypertension if they receive proper treatment. Nevertheless, in

Japan, only 12 million of 43 million individuals with hypertension receive treatment and have their BP controlled [1]. This phenomenon, termed the “hypertension paradox,” must be resolved to improve public health [2]. Among the reasons why individuals do not take action to control their hypertension,

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 1https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

the burden of time takes precedence for discontinuation or noninitiation of antihypertensive treatment. Telemedicine using internet-based communication may lower the hurdle for starting and adhering to hypertension treatment, eventually leading to the prevention of cardiovascular disease. This approach may also result in higher satisfaction among patients by allowing better use of their time rather than having them spend much of it waiting at clinics or hospitals. Technical difficulty and anxiety associated with telemedicine hence need to be managed sufficiently.

Achieving target BP levels in the treatment of hypertension requires patients’ adherence to and persistence in taking their medication. Self-measurement of home BP (SMBP) helps improve adherence to treatment and aids in BP control [3,4]. Adjusting antihypertensive medication based on self-measured home BP (HBP) for long periods is feasible, and HBP self-monitoring with self-titration of antihypertensive medication in accordance with an individualized predetermined protocol has been reported to result in better BP control than that with usual care (UC) [5].

In Japan, telemedicine without face-to-face communication has been permitted since early 2015. We developed an integrated web-based telemedicine platform to manage appointments, medical care, and payment without patients visiting a clinic. However, clinical evidence concerning the efficacy and safety of telemedicine remains scarce. In this study, we attempted to demonstrate the advantages of telemedicine over traditional care in the management of hypertension. The Paradigm of Antihypertensive Therapy along with Telemedicine Randomized (POATRAND) trial was designed and performed as a prospective, randomized, open-label, 2-arm study of patients with uncontrolled, uncomplicated hypertension to test the effectiveness and safety of BP telemonitoring as well as hypertension telemedicine.

Methods

Study Design and Participant Recruitment The POATRAND trial was a multicenter, open-label randomized controlled trial performed at Tokyo Women’s Medical University Clinic and private clinics in Japan. Potentially eligible participants with hypertension were recruited through web advertising. The inclusion criteria were age over 20 years, elevated systolic BP (SBP) or diastolic BP (DBP), ability to visit the Tokyo Women’s Medical University Clinic for the initial screening, willingness to receive hypertension care through telemedicine, and ability to self-measure HBP. The exclusion criteria were inability to use a smartphone, pregnancy, presence of major cardiovascular events and diabetes mellitus, an estimated glomerular filtration rate lower than 30

mL/min/1.73 m2 determined using the modified Modification of Diet in Renal Disease formula for Japanese patients [6], and presence of secondary hypertension excluding primary aldosteronism without surgical indication. The study protocol was approved by the Tokyo Women’s Medical University Research Ethics Committee (approval 160603) and registered in the UMIN Clinical Trials Registry under accession number

UMIN000025372. All patients provided their written informed consent to participate in the study.

Randomization Potentially eligible patients were invited to Tokyo Women’s Medical University Clinic between November 2015 and February 2017 for screening. After eligibility screening and provision of informed consent, each patient’s office BP was measured with the patient sitting quietly alone in a room. Before measurement, an experienced staff member instructed the patient on the procedure, placed a cuff with an appropriately sized bladder on the patient’s upper left arm, and left the room. BP was measured 3 times at 3-min intervals. Office BP was measured using the same validated BP monitor as the one used for home BP measurements (HEM-7252G-HP; Omron) [7], and the values were not concealed from the participants. At the second baseline visit, screening results were communicated to the participants, and the participants were stratified by the average of the second and third office SBP readings at the first visit and randomly assigned at a 1:1 ratio into the UC or telemedicine group, using an Excel-based random sampling number system.

Procedures HBP was measured using a 3G network–equipped automatic sphygmomanometer (HEM-7252G-HP; Omron) [7]. The device was based on the cuff–oscillometric principle and validated to meet the criteria of the Association for the Advancement of Medical Instrumentation. The device recorded and transmitted SBP and DBP values, heart rate, and the date and time of each measurement. Registered patients were instructed on how to use the device and asked to take their HBP reading in a sitting position twice every morning within 1 h of waking before taking a meal or medication and after more than 2 minutes of rest. Participants were also asked to measure their HBP twice every evening before going to bed. The telemedicine group used this device, consulted a physician through web-based visits in consideration of their transmitted BP values, and received prescription medication by mail for 1 year. The UC group used the same BP monitoring device but was managed with actual office visits using self-recorded BP readings, such as their diary entries. After randomization, baseline HBP was measured in both groups without changes in treatment. The standard visit-to-visit interval was set at 6 weeks for the telemedicine group and was unspecified for the UC group, leaving it to the physician’s discretion.

Primary and Secondary Outcomes The primary outcome was the average home SBP during the last week of the 12-month study period. Secondary outcomes included other measures of office BP and HBP including morning and evening SBP, DBP, and BP control rates, adverse events (eg, side effects and cardiovascular events), medication prescription (ie, number and defined daily dose), body weight, and laboratory measures.

Statistical Analysis We initially estimated that 260 patients were required for screening per group to detect a 4-mmHg difference in home SBP values between the 2 groups, with a 2-sided P value of .05

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 2https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

and 80% power. However, the study was prematurely terminated at the end of March 2018 because the health insurance policy in Japan changed, requiring a face-to-face visit at least every 3 months. SPSS (version 21, IBM Corp) was used for statistical analyses. All P values were 2-sided, and P≤.05 was considered statistically significant. Data are presented as mean (SD) values unless otherwise indicated.

Results

Baseline Participant Characteristics In total, 159 patients were screened for the study (Figure 1). Their baseline characteristics are shown in Table 1. The groups

did not show significant differences in age, female-to-male ratio, BMI, home SBP, home DBP, pulse rate, plasma aldosterone concentration, plasma renin activity, estimated glomerular filtration rate, hemoglobin A1c level, and low-density-lipoprotein cholesterol (LDL-C) level. At the end of the study, we assessed 46 individuals from the UC group and 48 from the telemedicine group. Dropout rates and adverse events are discussed subsequently.

Figure 1. CONSORT (Consolidated Standards of Reporting Trials) flowchart for the study. TM: telemedicine, UC: usual care.

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 3https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

Table 1. Baseline characteristics of the study participants (N=97).

Telemedicine group (n=49)Usual care group (n=48)All participantsParameters

28 (57%)30 (63%)58 (60%)Female, n (%)

53 (9)53 (9)53 (9)Age (years), mean (SD)

24.2 (3.4)24.3 (4.3)24.2 (3.9)BMI (kg/m2), mean (SD)

136 (15)136 (11)136 (13)Home systolic blood pressure (mmHg), mean (SD)

90 (10)91 (8)91 (9)Home diastolic blood pressure (mmHg), mean (SD)

73 (9)74 (9)73 (9)Home pulse rate (bpm), mean (SD)

26 (53)29 (60)55 (57)Antihypertensive treatment, n (%)

74.4 (14.4)77.3 (13.3)75.9 (14.0)Estimated glomerular filtration rate (mL/min/1.73 m2), mean (SD)

123.0 (28.3)119.2 (28.9)121.1 (28.8)Low-density-lipoprotein cholesterol (mg/dL), mean (SD)

5.7 (0.3)5.6 (0.3)5.7 (0.3)Hemoglobin A1c (%), mean (SD)

167 (136-232)150 (112-199)155 (119-216)Plasma aldosterone (pg/mL), median (IQR)

1.2 (0.9-2.0)0.9 (0.5-1.8)1.1 (0.6-1.9)Plasma renin activity (ng/mL/h), median (IQR)

Changes in HBP and Laboratory Data Before and After the Study Home SBP and DBP significantly decreased after 1 year in both groups (Tables 2 and 3). Plasma renin activity was significantly

increased in the UC group (P=.02) but not in the telemedicine group.

Table 2. Outcomes.

P valueTelemedicine groupUsual care group

P valuePostintervention (n=48)

Preintervention (n=49)

P valuePostintervention (n=46)

Preintervention (n=48)

.10.2923.9 (3.3)24.2 (3.4).863.9 (4.1)24.2 (4.3)BMI (kg/m2), mean (SD)

.02<.001125 (122-128)136 (132-140).01131 (128-134)136 (133-139)Home systolic blood pressure (mmHg), median (95% CI)

.07<.00183 (81-85)90 (87-93).0287 (85-89)91 (89-93)Home diastolic blood pressure (mmHg), median (95% CI)

.08.0871 (8)73 (9).4671 (7)74 (9)Home pulse rate (bpm), mean (SD)

.01<.00185.447.7<.00170.041.3Systolic blood pressure control rate (%)

.05.3072.4 (14.4)74.4 (14.4).1676.8 (11.9)77.3 (13.3)Estimated glomerular filtration rate

(mL/min/1.73 m2), mean (SD)

.93.654.1 (0.3)4.1 (0.3).814.1 (0.3)4.1 (0.3)Potassium (mEq/L), mean (SD)

.04.11123.8 (28.6)123.0 (28.3).53110.3 (22.7)119.2 (28.9)Low-density-lipoprotein choles- terol (mg/dL), mean (SD)

.53.605.7 (0.4)5.7 (0.3).745.7 (0.3)5.6 (0.3)Hemoglobin A1c (%), mean (SD)

.49.54171 (139-229)167 (136-232).07151 (115-226)150 (112-199)Plasma aldosterone (pg/mL), medi- an (IQR)

.57.071.9 (1.1-4.8)1.2 (0.9-2.0).021.1 (0.6-3.6)0.9 (0.5-1.8)Plasma renin activity (ng/mL/h), median (IQR)

Table 3. Home blood pressure change from baseline till the end of the 1-year study period.

P valueTelemedicine groupUsual care group

.23–9.2 (14.3)–5.4 (11.3)Change in systolic blood pressure (mmHg), mean (SD)

.33–5.5 (8.7)–3.5 (8.1)Change in diastolic blood pressure (mmHg), mean (SD)

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 4https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

Differences in HBP Between the Participant Groups The average home SBP during the last week of the study was significantly lower (by 6 mmHg) in the telemedicine group than in the UC group (Table 2). Home DBP after the 1-year study period tended to be lower in the telemedicine group, but the difference was not significant. The average SBP and DBP reached the therapeutic targets of less than 135 mmHg and 85 mmHg, respectively, at the time of measurement in the telemedicine group only.

When morning (4-10:59 AM) and evening (6 PM to 3:59 AM) BPs were analyzed separately (Table 4), the telemedicine group showed significantly lower evening SBP and DBP readings. The average morning SBP and DBP readings were also lower in the telemedicine group, but the difference was not significant.

The number of BP measurements per week for the whole study period was significantly higher in the telemedicine group (17.8, SD 11.5) than in the UC group (12.1, SD 11.0) (P=.02).

Table 4. Average morning and evening home blood pressure readings during the last week of the 1-year study period.

P valueTelemedicine group (n=48), mean (SD)

Usual care group (n=46), mean (SD)

Morning

.09130.6 (10.3)134.0 (8.6)Home systolic blood pressure (mmHg)

.1488.3 (7.7)90.6 (7.2)Home diastolic blood pressure (mmHg)

.2569.7 (7.5)71.6 (8.0)Home pulse rate (bpm)

Evening

.007125.8 (11.5)131.6 (8.7)Home systolic blood pressure (mmHg)

.00382.5 (7.5)87.2 (7.5)Home diastolic blood pressure (mmHg)

.2774.2 (8.5)76.3 (9.4)Home pulse rate (bpm)

Clinical Parameters at the End of the Study At the end of the study, LDL-C was significantly lower in the UC group than in the telemedicine group. Plasma renin activity was not significantly different between the 2 groups at baseline, but significantly increased at the endpoint only in the UC group. The endpoint plasma renin activity was not significantly

different between the 2 groups. Other laboratory data were not significantly different between the 2 groups.

Prescription Data for Antihypertensive Medications Percentages of antihypertensive treatment are shown in Table 5.

Table 5. Prescription data.

Telemedicine groupUsual care group

Postintervention (n=48), n (%)Preintervention (n=49), n (%)Preintervention (n=48), n (%)

7 (14.6)26 (54.2)29 (60.0)No medication

8 (16.7)13 (27.1)11 (23.0)Calcium channel blocker only

3 (6.3)2 (4.2)0 (0)Angiotensin II receptor blocker only

3 (6.3)0 (0)2 (4.0)Mineralocorticoid receptor blocker only

22 (45.8)7 (14.6)5 (10.0)Angiotensin II receptor blocker/angiotensin convert- ing enzyme inhibitor + calcium channel blocker

3 (6.3)1 (2.1)1 (2.0)Mineralocorticoid receptor blocker + calcium channel blocker

0 (0)0 (0)0 (0)Angiotensin II receptor blocker + diuretic

2 (4.2)0 (0)0 (0)Angiotensin II receptor blocker/angiotensin convert- ing enzyme inhibitor + calcium channel blocker + diuretic

Dropout Rates and Adverse Events Of the 50 and 49 participants allocated to the UC and telemedicine groups, respectively, we assessed 46 from the UC group and 48 from the telemedicine group. In the UC group, a participant requested telemedicine, another did not meet the inclusion criteria, and another was lost to follow-up; hence, all 3 were excluded from the study. One participant from the UC

group experienced a mild subarachnoid hemorrhage with no neurological deficits or hospitalization and dropped out of the study. In the telemedicine group, 1 participant experienced angina pectoris and discontinued the intervention. Medication-related complaints upon initiation or change of antihypertensive drugs included urticaria (n=1) and concerns of having considerably low BP (n=3). No discontinuation owing

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 5https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

to drug side effects or difficulty using the telemedicine interface was recorded.

Discussion

Principal Findings This study, for the first time in Japan, conducted hypertension treatment using telemonitoring and telemedicine without face-to-face visits for 1 year and revealed 2 major findings. First, telemedicine without actual office visits was determined to be relatively safe in managing hypertension for 1 year. Second, the telemedicine group achieved a lower BP than the UC group. Although the BP difference from baseline was not significantly different between the groups, the telemedicine group demonstrated a reduction in SBP of 9.2 mmHg, whereas the reduction in the UC group was 5.4 mmHg.

In a previous study that investigated the safety of telemedicine without office visits, the number of adverse events was not significantly different from that of UC. In the Telemonitoring and Self-management of Hypertension (TASMINH2) Trial, the frequency of side effects, such as swelling of legs, stiff joints, fatigue, and cough, was similar between the telemedicine and UC groups [5]. In our study, 2 patients dropped out owing to cardiovascular events during the study: 1 from the telemedicine group for new-onset angina pectoris and 1 from the UC group for subarachnoid hemorrhage. No significant difference in the laboratory findings was observed between the 2 groups at the end of the study, except for lower LDL-C in the UC group. Medication-related complaints were successfully managed through web-based consultations. The results of this and previous studies suggest that telemedicine is reasonably safe for use in controlling BP in uncomplicated hypertension.

Superior BP control in the telemedicine group than that in the UC group could be attributed to several factors, one being the intensity of the intervention. Sheppard et al [8] demonstrated that intense interventions, such as pharmacotherapeutic intervention managed by a pharmacist, with frequent telemonitoring, were more effective than low-intensity interventions, such as telemonitoring only, in patients with obesity. In this study, the telemedicine group received team-based care from physicians specializing in hypertension, and the participants were able to ask questions using the app. Therefore, the effects of antihypertensive intervention could have been enhanced by the more intensive interventions and greater expertise that the telemedicine group received. Several studies found that telemonitoring of BP improves BP control [3,4]. On the other hand, other studies showed that telemonitoring by itself does not significantly improve it [9]. Pellegrini et al [10] recently reported that self-monitoring of BP at home with co-interventions, such as telecounseling, favorably controlled BP more than self-monitoring of BP alone did. Nevertheless, regardless of their effects on BP control, advances in internet technology and data technology should accelerate the dissemination of BP telemonitoring and

telemedicine, which would make retrospective analysis of BP management interventions in the future much easier than those in the past, when such analysis depended on a paper-based approach. In our previous study, an electronic sphygmomanometer with an automated 3G data transfer and room temperature monitor was used for daily SMBP [11]. BP variability based on SMBP can predict cardiovascular outcome in patients with hypertension [12,13]. Room temperature at the time of SMBP significantly correlates with variability of SBP [14,15]. Therefore, an HBP monitor equipped with various sensors, serving as the “Internet of Things,” may be used to monitor environmental conditions to maintain the optimal BP in future studies.

Regarding adherence, the frequency of HBP measurements was much higher in the telemedicine group than in the UC group. We do not have data on medication adherence, but Ogedegbe and Schoenthaler [16] demonstrated that HBP self-monitoring significantly improved medication adherence and reported that the telemedicine group self-monitored their BP more frequently.

Limitations This study has several limitations. First, this study was limited by its design in that the physicians’ level of expertise differed for the telemedicine and UC groups. Second, although the rate of treated hypertension at baseline was not different between the groups, changes in prescription were precisely tracked only in the telemedicine group. Third, the number of visits for the UC group was not determined, and the average number of visits could not be compared. In Japan, regular face-to-face visits for hypertension management in clinics generally occur at 2- to 8-week intervals. Fourth, because the study was terminated prematurely, it did not reach its intended sample size, making the power of the study insufficient for some analyses. Nevertheless, our finding of a significantly lower HBP at the end of the study period in the telemedicine group despite a smaller-than-intended sample size is promising for the improvement of BP control through telemedicine. Fifth (related to the first limitation), although the participants in the UC group were referred to primary care physicians with a letter asking them to target an HBP of less than 135/85 mmHg, it is not clear whether the physicians actually adhered to this request [17]. Although this study cannot delineate the effects of telemonitoring, telemedicine, and specialist intervention compared with UC, ours is a real-world study providing pilot data on hypertension telemedicine in Japan, which can be used to design future studies.

Conclusions Our results suggest that antihypertensive telemedicine using HBP telemonitoring and web-based video visits is safe. The telemedicine group of patients with uncomplicated hypertension achieved better BP control than the group assigned to conventional care. Further investigations are required to elucidate the benefits of telemedicine in treating hypertension on a larger scale.

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 6https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

Acknowledgments This study was conducted at Tokyo Women’s Medical University. This study was supported in part by the PORT Corporation. We gratefully acknowledge the excellent technical assistance of Erina Yamaguchi, who participated in data collection and analysis.

Authors' Contributions JY and MSY contributed to the design of the study and performed data collection and analysis. JY, MSY, RO, and AI drafted, critically revised, and approved the final version of the paper for submission.

Conflicts of Interest JY and MSY are directors of General Incorporated Association TelemedEASE, which aims to facilitate safe and effective telemedicine.

Multimedia Appendix 1 CONSORT-eHEALTH checklist (V 1.6.1). [PDF File (Adobe PDF File), 729 KB-Multimedia Appendix 1]

References

1. Umemura S, Arima H, Arima S, Asayama K, Dohi Y, Hirooka Y, et al. The Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH 2019). Hypertens Res 2019 Sep;42(9):1235-1481. [doi: 10.1038/s41440-019-0284-9] [Medline: 31375757]

2. Chobanian AV. Shattuck Lecture. The hypertension paradox–more uncontrolled disease despite improved therapy. N Engl J Med 2009 Aug 27;361(9):878-887. [doi: 10.1056/NEJMsa0903829] [Medline: 19710486]

3. Duan Y, Xie Z, Dong F, Wu Z, Lin Z, Sun N, et al. Effectiveness of home blood pressure telemonitoring: a systematic review and meta-analysis of randomised controlled studies. J Hum Hypertens 2017 Jul;31(7):427-437. [doi: 10.1038/jhh.2016.99] [Medline: 28332506]

4. Omboni S, Gazzola T, Carabelli G, Parati G. Clinical usefulness and cost effectiveness of home blood pressure telemonitoring: meta-analysis of randomized controlled studies. J Hypertens 2013 Mar;31(3):455-467. [doi: 10.1097/HJH.0b013e32835ca8dd] [Medline: 23299557]

5. McManus R, Mant J, Bray E, Holder R, Jones MI, Greenfield S, et al. Telemonitoring and self-management in the control of hypertension (TASMINH2): a randomised controlled trial. Lancet 2010 Jul 17;376(9736):163-172. [doi: 10.1016/S0140-6736(10)60964-6] [Medline: 20619448]

6. Horio M, Imai E, Yasuda Y, Watanabe T, Matsuo S. Modification of the CKD epidemiology collaboration (CKD-EPI) equation for Japanese: accuracy and use for population estimates. Am J Kidney Dis 2010 Jul;56(1):32-38. [doi: 10.1053/j.ajkd.2010.02.344] [Medline: 20416999]

7. Takahashi H, Yoshika M, Yokoi T. Validation of two automatic devices: Omron HEM-7252G-HP and Omron HEM-7251G for self-measurement of blood pressure according to the European Society of Hypertension International Protocol revision 2010. Blood Press Monit 2015 Oct;20(5):286-290. [doi: 10.1097/MBP.0000000000000127] [Medline: 25932887]

8. Sheppard J, Tucker K, Davison W, Stevens R, Aekplakorn W, Bosworth HB, et al. Self-monitoring of Blood Pressure in Patients With Hypertension-Related Multi-morbidity: Systematic Review and Individual Patient Data Meta-analysis. Am J Hypertens 2020 Mar 13;33(3):243-251 [FREE Full text] [doi: 10.1093/ajh/hpz182] [Medline: 31730171]

9. Tucker KL, Sheppard JP, Stevens R, Bosworth HB, Bove A, Bray EP, et al. Self-monitoring of blood pressure in hypertension: A systematic review and individual patient data meta-analysis. PLoS Med 2017 Sep;14(9):e1002389 [FREE Full text] [doi: 10.1371/journal.pmed.1002389] [Medline: 28926573]

10. Pellegrini D, Torlasco C, Ochoa JE, Parati G. Contribution of telemedicine and information technology to hypertension control. Hypertens Res 2020 Jul;43(7):621-628. [doi: 10.1038/s41440-020-0422-4] [Medline: 32203451]

11. Yatabe J, Yatabe MS, Morimoto S, Watanabe T, Ichihara A. Effects of room temperature on home blood pressure variations: findings from a long-term observational study in Aizumisato Town. Hypertens Res 2017 Aug;40(8):785-787. [doi: 10.1038/hr.2017.18] [Medline: 28230196]

12. Stevens SL, Wood S, Koshiaris C, Law K, Glasziou P, Stevens RJ, et al. Blood pressure variability and cardiovascular disease: systematic review and meta-analysis. BMJ 2016 Aug 09;354:i4098 [FREE Full text] [doi: 10.1136/bmj.i4098] [Medline: 27511067]

13. Juhanoja EP, Niiranen TJ, Johansson JK, Puukka PJ, Thijs L, Asayama K, International Database on Home Blood Pressure in Relation to Cardiovascular Outcome (IDHOCO) Investigators. Outcome-Driven Thresholds for Increased Home Blood Pressure Variability. Hypertension 2017 Apr;69(4):599-607. [doi: 10.1161/HYPERTENSIONAHA.116.08603] [Medline: 28193705]

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 7https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX

14. Iwahori T, Miura K, Obayashi K, Ohkubo T, Nakajima H, Shiga T, et al. Seasonal variation in home blood pressure: findings from nationwide web-based monitoring in Japan. BMJ Open 2018 Jan 05;8(1):e017351 [FREE Full text] [doi: 10.1136/bmjopen-2017-017351] [Medline: 29306878]

15. Hanazawa T, Asayama K, Watabe D, Tanabe A, Satoh M, Inoue R, HOMED‐BP (Hypertension Objective Treatment Based on Measurement by Electrical Devices of Blood Pressure) Investigators. Association Between Amplitude of Seasonal Variation in Self-Measured Home Blood Pressure and Cardiovascular Outcomes: HOMED-BP (Hypertension Objective Treatment Based on Measurement By Electrical Devices of Blood Pressure) Study. J Am Heart Assoc 2018 May 04;7(10):e008509 [FREE Full text] [doi: 10.1161/JAHA.117.008509] [Medline: 29728372]

16. Ogedegbe G, Schoenthaler A. A systematic review of the effects of home blood pressure monitoring on medication adherence. J Clin Hypertens (Greenwich) 2006 Mar;8(3):174-180 [FREE Full text] [doi: 10.1111/j.1524-6175.2006.04872.x] [Medline: 16522994]

17. Nishigaki N, Shimasaki Y, Yoshida T, Hasebe N. Physician and patient perspectives on hypertension management and factors associated with lifestyle modifications in Japan: results from an online survey. Hypertens Res 2020 May;43(5):450-462 [FREE Full text] [doi: 10.1038/s41440-020-0398-0] [Medline: 31996815]

Abbreviations BP: blood pressure DBP: diastolic blood pressure HBP: home blood pressure LDL-C: low-density-lipoprotein cholesterol POATRAND: Paradigm of Antihypertensive Therapy Along With Telemedicine Randomized Trial SBP: systolic blood pressure SMBP: self-measurement of home blood pressure UC: usual care

Edited by G Eysenbach; submitted 31.01.21; peer-reviewed by B Green, S Omboni; comments to author 22.02.21; revised version received 20.07.21; accepted 28.07.21; published 31.08.21

Please cite as: Yatabe J, Yatabe MS, Okada R, Ichihara A Efficacy of Telemedicine in Hypertension Care Through Home Blood Pressure Monitoring and Videoconferencing: Randomized Controlled Trial JMIR Cardio 2021;5(2):e27347 URL: https://cardio.jmir.org/2021/2/e27347 doi: 10.2196/27347 PMID: 34321194

©Junichi Yatabe, Midori Sasaki Yatabe, Rika Okada, Atsuhiro Ichihara. Originally published in JMIR Cardio (https://cardio.jmir.org), 31.08.2021. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Cardio, is properly cited. The complete bibliographic information, a link to the original publication on https://cardio.jmir.org, as well as this copyright and license information must be included.

JMIR Cardio 2021 | vol. 5 | iss. 2 | e27347 | p. 8https://cardio.jmir.org/2021/2/e27347 (page number not for citation purposes)

Yatabe et alJMIR CARDIO

XSL•FO RenderX