Telemedicine in Southeast Asia: A Systematic Review

Abstract

Introduction:

Telemedicine is an effective means of delivering health care in Southeast Asian (SEA) countries. This systematic review explored the use of telemedicine systems for delivering health care services in SEA countries during the past 10 years.

Methods:

Literature searches were conducted in PubMed, Embase, Medline, Global Health, and CENTRAL (Cochrane Central Register of Controlled Trials). The inclusion criteria were as follows: (1) studies published between 2010 and 2021; (2) study settings located in SEA countries; (3) articles published in English; and (4) availability of a full-text version of the article. Information was extracted and evaluated for each study based on quality and risk of bias.

Results:

Thirty-seven of 6,554 records were eligible for inclusion. Studies included based on percentage were descriptive (29.73%), cost-effective (8.11%), randomized controlled trials (5.41%), and mixed methods (2.7%). Teleophthalmology and teleconsultation were the major reasons for using telemedicine, representing 21.62% of all studies. A hub-and-spoke and store-and-forward models were used. Free messenger applications supported communication modalities in or out of the systems.

Discussion:

The COVID-19 (coronavirus disease 2019) pandemic increased research studies on telemedicine, with most studies occurring in Singaporean hospitals (49%). Descriptive studies predominated, followed by retrospective and cross-sectional studies.

Conclusions:

Our findings demonstrated that telemedicine was a powerful tool. It is feasible, safe, effective, and less expensive than traditional methods. However, robust research is needed to fully investigate telemedicine systems in SEA countries.

Introduction

Telemedicine is often defined as the delivery of health care service at a distance using information and communication technology (ICT) by clinicians and health care professionals for diagnosis, treatment, prevention of disease or injuries, research, education, evaluation, and the exchange of valid information.¹ Technologies used in telemedicine are diverse in nature and complexity.² The Center for Connected Health Policy (CCHP) categorized telemedicine into four types: live video, store and forward, remote patient monitoring, and email/phone/fax.³

Telemedicine is a useful tool that provides an efficient and cost-effective way to deliver health care.⁴ It has been used for more than 100 years to help health care workers deliver services. In 1906, Willem Einthoven, a Dutch physiologist who developed a technique similar to telemedicine, created the first mobile electrocardiograph device to record the electrical cardiac signals of patients from a hospital 1.5 km away from his laboratory.⁵ The main benefits of telemedicine are reduced travel costs and time saved, especially for people who live in rural areas with limited access to health care services.⁶ Telemedicine provides good opportunities for emerging and developing countries to access basic primary health care services. Health care providers are able to evaluate, diagnose, treat, and provide follow-up care to distant patients by using telemedicine applications.^7,8

Telemedicine effectively delivers last mile health care in Southeast Asian (SEA) countries. ACCESS Health International Southeast Asia recently reported that digital technology is the key to achieving universal health coverage in the Association of Southeast Asian Nations (ASEAN).⁹ In Thailand, an initiative, which included the Ministry of Public Health and the National Broadcasting and Telecommunications Commission, used telemedicine to increase health care access for remote areas via telemedicine systems.^10,11

In 2012, the national Expanded Program on Immunization in Vietnam developed and piloted a digital immunization registry program known as ImmReg, a web-based application system to register new clients, to track due dates, and to identify doses required for on-time vaccination. The system automatically sends short message service (SMS) reminders to clients and generates reports to health care workers.¹²

In the past decade, many ASEAN countries implemented telemedicine services, but there were no national scientific institutions whose focus was to develop telemedicine on a broad scale. Rather, most efforts were short-term demonstration projects, carried out in conjunction with international companies and large hospitals or universities.¹³ Furthermore, there was little published evidence¹⁴ to document the existence of the programs or to evaluate the quality, accessibility, effectiveness, and cost of the telemedicine care provided.¹⁵ Telehealth was not integrated into existing health care service systems because of the financial and technological barriers. In addition, physicians' lack of expertise with the technology also served as impediments to the implementation of telemedicine programs.^13,15,16

During the coronavirus disease 2019 (COVID-19) pandemic, policy changes reduced restrictions on telemedicine, thus enabling health care entities to provide services when face-to-face care was neither safe nor practical.^17,18 Health care providers and systems promptly adopted and expanded telemedicine services to provide a safe online service in place of in-person visits.¹⁹ Unfortunately, telemedicine has some limitations, for instance, it limits the ability of the health care professional to thoroughly observe the patient. The lack of in person contact may negatively impact the guidance provided to the patient and the clinical decisions made by the health care professional. The objective of this systematic review was to explore the implementation and use of telemedicine systems for delivering health care services in SEA countries during the past 10 years.

Methods

A systematic review of the literature was conducted to elucidate the current use and implementation of clinical care provided via telemedicine in SEA countries.¹

TYPE OF STUDY

The systematic review included: descriptive, observational, analytical, quasi-experimental, clinical trial, and randomized controlled trial studies. Studies that reported on patient outcomes, evaluation methods, or educational outcomes were also included in the review. Surveys or interviews of experts' opinion or experience were excluded.

INFORMATION SOURCES

Literature searches were conducted in PubMed, Embase, Medline, Global Health, and Cochrane Central Register of Controlled Trials (CENTRAL) up to April 25, 2021. Selected reference lists from related articles were also screened to identify additional appropriate studies. Medical subject headings (MeSH) and keywords were used to search PubMed, Embase, and Medline. Relevant articles were retrieved and imported to a reference manager software program, and duplicates were deleted.

SEARCH STRATEGY

Telemedicine and Southeast Asia were the main MeSH terms used to identify relevant studies. In addition to MeSH headings, names and synonyms were used to retrieve the greatest number of articles related to telemedicine in all SEA countries. The search terms used in each electronic database are presented in Table 1.

Table 1.

Literature Searches and Strategies

DATABASES	DATE OF SEARCH	SEARCH STRATEGIES	RESULTS
PubMed	April 25, 2021	((“Telemedicine”[Mesh]) OR (((telemedicine[Title/Abstract]) OR (telehealth[Title/Abstract])))) AND ((((“Asia, Southeastern”[Mesh]) OR (((southeast asia[Title/Abstract]) OR (southeastern asia[Title/Abstract])))) OR (“Australia”[Mesh])) OR (Australia[Title/Abstract]))	1,504
Embase	April 25, 2021	#7 #3 AND #6 #6 #4 OR #5 #5 ‘southeast asia’/exp OR ‘australia’/exp #4 ‘southeast asia’:ab,ti OR australia:ab,ti #3 #1 OR #2 #2 ‘telemedicine’/exp OR ‘telehealth’/exp #1 telemedicine:ab,ti OR telehealth:ab,ti	2,158
Medline (Ovid) (1946–April 23, 2021)	April 25, 2021	1 (telemedicine or telehealth).af. 2 (telemedicine or telehealth).sh. 3 1 or 2 4 (southeast asia or australia).af. 5 (southeast asia or australia).sh. 6 4 or 5 7 3 and 6	2,486
Global Health (1973–2021 week 16)	April 25, 2021	1 (telemedicine or telehealth).af 2 (telemedicine or telehealth).sh. 3 1 or 2 4 (southeast asia or australia).af. 5 (southeast asia or australia).sh. 6 4 or 5 7 3 and 6	278
CENTRAL (March 2021)	April 25, 2021	1 (telemedicine or telehealth).ab 2 (telemedicine or telehealth).sh. 3 1 or 2 4 (southeast asia or australia).ab. 5 (southeast asia or australia).sh. 6 4 or 5 7 3 and 6	111

CENTRAL, Cochrane Central Register of Controlled Trials.

STUDY INCLUSION AND EXCLUSION CRITERIA

The inclusion criteria were as follows: (1) studies published between 2010 and 2021; (2) study settings located in SEA countries, including Brunei, Burma (Myanmar), Cambodia, Timor-Leste, Indonesia, Laos, Malaysia, the Philippines, Singapore, Thailand, and Vietnam. The MeSH heading Australia was used to retrieve additional articles since the country provides support to SEA nations in the health informatics domain; (3) articles published in English; and (4) a full-text version of the article was available. The exclusion criteria were e-health or m-health interventions, survey, and review articles. Endnote was used to retrieve and save research records and articles.²⁰

STUDY SELECTION

Relevant records were retrieved and imported to a reference manager software program, and duplicates were deleted. All records were screened and evaluated for compliance with the inclusion and exclusion criteria. Full-text versions of the eligible records were retrieved and read.

DATA EXTRACTION

A data extraction form to record the general characteristics, use, and implementation of included studies was adaptively created based on the guideline for developing a data extraction from the Cochrane Handbook for Systematic Reviews of Interventions.²¹ For each included study, the following information was recorded: last name of the first author, year of publication, study design, sample size, duration of study, location and country of study, type of telemedicine, participant and disorder, organizational setting, system and technology used, implementation, and main finding of study. A.P. extracted the data from included studies, and then, D.G.S. checked and evaluated the extracted data.

QUALITY OF STUDIES

Each eligible study was evaluated for quality and risk of bias based on the Statement on reporting of evaluation of studies in Health Informatics (STARE-HI).²² Study design, theoretical background, participant, study flow, outcome measures or evaluation criteria, methods for data acquisition and measurement, and methods for data analysis were all included. The Cochrane Assessment Risk of Bias tool was used to analyze and present the risk of bias graphs.²³

STATISTICAL ANALYSIS

Descriptive statistics were performed for general characteristics of included studies, including frequency and percentage. All information regarding telemedicine is presented.

Results

The literature search identified 6,554 records. After 2,417 duplicates were removed, the titles and abstracts of the unique 4,137 records were screened by applying the inclusion and exclusion criteria for removing extraneous titles. All remaining records were downloaded in full-text format, and the articles were screened again based on the inclusion and exclusion criteria. Thirty-seven eligible studies were included in the review (Fig. 1).

Fig. 1.

Study flow diagram.

Singapore (48.65%) was the most frequently recorded study setting, followed by Thailand (21.62%) and Malaysia (10.81%). Various study designs were used to investigate telemedicine services, including descriptive studies (29.73%), cost-effective studies (8.11%), randomized controlled trials (5.41%), and mixed methods studies (2.7%). Teleophthalmology and teleconsultation were the major reasons for using telemedicine, representing 21.62% of all studies (Table 2). Most of the included studies were published in 2020 (32.48%), followed by 2015 (10.80%) and 2013 (10.79%) (Fig. 2).

Fig. 2.

Publication years of included studies.

Table 2.

Characteristics of Included Studies

CHARACTERISTIC	n (%)
Country
1. Singapore	18 (48.65)
2. Thailand	8 (21.62)
3. Malaysia	4 (10.81)
4. Lao People's Democratic Republic	2 (5.41)
5. Indonesia	2 (5.41)
6. Cambodia	1 (2.7)
7. Vietnam	1 (2.7)
8. Philippines	1 (2.7)
Study design
1. Descriptive study	11 (29.73)
2. Retrospective study	8 (21.62)
3. Cross-sectional study	5 (13.51)
4. Prospective study	3 (8.11)
5. Cost-effective study	3 (8.11)
6. Observational study	2 (5.41)
7. Randomized control trial	2 (5.41)
8. Clinical trial	1 (2.7)
9. Mixed method study	1 (2.7)
10. Commentary	1 (2.7)
Type of telemedicine
1. Teleophthalmology	8 (21.62)
2. Teleconsultation	8 (21.62)
3. Telehealth	5 (13.51)
4. Telestroke	4 (10.81)
5. Telemonitoring	3 (8.11)
6. Teleneurosurgery	2 (5.41)
7. Teleradiology	2 (5.41)
8. Telepathology	1 (2.7)
9. Telegraphic medicine	1 (2.7)
10. Teledermatology	1 (2.7)
11. Telerehabilitation	1 (2.7)
12. Telegeriatric	1 (2.7)

IMPLEMENTATION AND USE OF TELEMEDICINE

All included studies established that telemedicine was used in several organizational settings, and different types of telemedicine were used for different clinical conditions. We summarized the details of each study (Table 3).

Table 3.

Implementation and Use of Telemedicine of Included Studies

STUDY	METHOD AND DESIGN	PARTICIPANTAND CONDITIONS	ORGANIZATIONAL SETTING	TYPE OF TELEMEDICINE	SYSTEM AND TECHNOLOGY	MAIN FINDINGS	ISSUES AND OTHER FINDINGS
1. Muengtaweepongsa⁷³	Descriptive study	Patients with acute stroke	Thammasat Stroke Center, Thailand	Telestroke (teleconsultation, teleradiology)	A CT digital imaging and communication transferred directly to the stroke consultants. Patients' information was communicated to the stroke consultants by usual telephone discussion.	Partial teleconsultation with remote radiology interpretation alone was feasible and safe for administration of rt-PA to acute stroke patients.	Implementation of complete teleconsultation with both video-teleconferencing and teleradiology is expensive.
2. Narenthiranathan⁶⁴	Descriptive study	Referred neurosurgery cases	Hospital Kuala Lumpur, Malaysia	Teleneurosurgery (teleconsultation)	Referral, teleneurosurgery, system with PACS for patient registration, online clinical data, load image (CT, MRI), send and receive data between physician and neurosurgeon Alert medical doctors by call, SMS (confirm received)	681 of 970 referrals were made by teleneurosurgery system with 291 referrals via email and 12 referrals via MMS.	Low-quality images were sent by MMS. Telephone advice may have led to misinterpretations. Systems were down.
3. Ausayakhun⁷⁴	Cross-sectional study	CMV retinitis among HIV patients	A university-based tertiary medical center Chiang Mai, Thailand	Telemedicine for CMV retinitis diagnosis	A trained ophthalmic photographer used a digital fundus camera to capture both individuals and composite images of eyes. Images were transferred to a secure online server via a file transfer protocol from on-site to ophthalmic clinic, and images were graded and diagnosed blindly by three experts.	The sensitivity for detecting CMV retinitis by distant graders using composite retinal pictures ranged from 89% to 91%, but the specificity ranged from 85% to 88%.	The feasibility of telemedicine for CMV retinitis diagnosis is dependent on the ease of use of the imaging system with a retinal camera. Image blur, duplication, and improper image placement.
4. Ang²⁴	Retrospective observational study	Acute ischemic stroke patients	A 750-bed hospital in eastern Singapore	Telestroke	The system is based on a “hub-and-stroke” concept. An acute stroke patient at the ED of study hospital (spoke) can be examined by the duty neurologist at the NNI (hub) via a high-definition webcam. The neurologist can view the image of the patient's CT brain scan via high-speed internet connection. ED physician can assist and communicate in real time with the neurologist until a treatment decision of thrombolysis is made.	The telestroke system provided a longer mean DTN time (92.7 vs. 68.1 min), a shorter mean DTD (92.7 vs. 68.1 min) compared with an original stroke protocol system. The onset of symptoms to needle time was 154.9 min.	The additional steps of logging onto the telemedicine system and establishing contact with the neurologist. Staff member might not be familiar with the use of the telemedicine equipment. Junior ED doctors had a longer mean DTN time than Senior ED doctors.
5. Kadaba²⁶	Retrospective analysis study	Not specified (histopathology)	The CSC, Phnom Penh, Cambodia	Telepathology	A hybrid telepathology network, based on store-and-forward communication using the iPATH network and videoconferencing using Skype. Tissue slides are prepared at a Cambodian laboratory. Images are captured and uploaded to the iPATH network along with diagnostic comments. Specialists worldwide who have been invited to join the group in the iPATH network are notified and provided second opinions on the uploaded cases.	38 cases were uploaded to the iPATH network for additional telemedicine discussion during the 8-month period of the study. 3 of these cases were no responses from specialists on iPATH. 7 clinical videoconferences were held with specialists globally for final diagnosis and management plans.	Patients are easily lost to follow-up or difficult to track if they are transferred to other hospitals, so obtaining complete patients' outcome data are challenging. Operating a telepathology link presents logical challenges and establishing a network and organizing conferences required appropriate technical resources, support and a committed staff willing to attend meetings.
6. Shah⁷⁵	Descriptive study	CMV retinitis among HIV patients	Tan Tock Seng Hospital, Singapore	Telemedicine for CMV retinitis screening	A qualified technician used a nine-filed DFP to capture patient eyes and reconstruct the nine images into one composite montage image. The composite image was evaluated without the patient's history by a retinal specialist.	DFP was an effective approach of screening for CMV with 100% sensitivity and 99.9% compared with indirect ophthalmoscopy.	Low quality of photographs or unreadable images.
7. Tan³⁶	Prospective study	Chronic blurring of vision	A primary eye care facility, Singapore TTSH, Singapore	Teleophthalmology	Patient's information included captured images and videos from cameras attached to the slit lamp. The information were stored and transferred securely from the primary care facility to the TTSH. Videoconferencing between the patient, the primary care facility, and ophthalmologist at TTSH was performed with a shared electronic whiteboard after transmission of data.	Teleophthalmology had 96% sensitivity and 100% specificity to identify visually significant media opacity. 97% of patients were satisfied with the system's process.	Low quality of the fundus images.
8. Maamari⁷⁶	Descriptive study	Corneal epithelial defects	Chiang Mai University Hospital, Chiang Mai, Thailand	Telemedicine for diagnosis of corneal abrasions and ulcers	A smartphone (iPhone 4s) attachment with a +25-diopter lens and external LED light supplies was employed to deliver magnified images of the cornea with controlled illumination. On-site ophthalmologist was connected to three off-site ophthalmologists.	The agreement of an epithelial defect between an on-site ophthalmologist and the off-site ophthalmologists was good (0.73) with 83.8% of sensitivity and 97% specificity.	N/A
9. Wai Leng⁴⁴	Observational study	Patients with uncontrolled diabetes	CGH, a 790-bed regional hospital, Singapore	Telehealth	Telecarers used a telephone to call participants once a month for the first 3 months to provide personalized health education modules on diabetes, cholesterol, and blood pressure management and recognition of symptoms. The telecarers made a follow-up visit 3 months after the last call to monitor and support the patients.	The telephone support led to a greater reduction in HbA1c among patients with higher baseline level (HbA1c 58%).	N/A
10. Jirawison⁷⁷	Clinical trial	CMV retinitis among HIV patients	Nakornping Hospital, a tertiary medical center, Chiang Mai, Thailand	Telemedicine screening for CMV retinitis	Researchers captured mosaic fundus photographs through a dilated pupil using a digital fundus camera. An expert on-site ophthalmologist screened the results of the fundus images and then assessed each eye with indirect ophthalmoscopy and noted the clinical findings on the form. The mosaic image for each eye was uploaded to a secure server and assigned to three distant graders who evaluated each image for CMV retinitis.	The average sensitivity for the three distant graders to detect CMV retinitis on fundus images was 30.2%, whereas the average specificity was 99.1%.	Patients did not want to undergo pupil dilation because they were concerned about driving home after the procedure. Clinical staff often did not have time to obtain informed consent on busy days. Patients without visual symptoms did not want to spend the extra time needed to participate in the study.
11. Lee⁷⁸	Mixed method study	Muslims with type 2 diabetes who were fasting during Ramadan	Malaysia	Telemonitoring	All participants had a pre-Ramadan educational seminar about management of their diabetes. Telemonitoring group received a web-enabled glucometer that enabled self-monitoring and goal setting. Recorded readings were transmitted to an online portal. Focus group was conducted with participants to determine their experience and attitude during the program.	The telemonitoring patients were less likely to experience hypoglycemia than the control group (odds ratio: 0.13), and there were no differences in glycemic control.	Participants were concerned about technological barriers, but they viewed telemedicine as a more convenient tool. Records based on self-reports may not be entirely reliable. Difficulty in accessing an internet connection. Participants' lack of knowledge in using technology. A participant was afraid because he believed that he was being monitored by others.
12. Matsumoto⁷⁹	Descriptive study	Patients who underwent PR for assessment of periodontitis, dental caries, inflammatory disease, and/or temporomandibular disorders	Faculty of Dentistry, the UHS, Lao People's Democratic Republic Department of OMFR, Nihon University School of Dentistry, Japan	Teleradiology	Multiple dentists conducted PR scanning using reference planes. All radiographs were stored, and images were sent from the UHS Faculty of Dentistry to the Nihon University School of Dentistry Department of OMFR through the teleradiology system. A Japanese OMFR specialist performed triage of patient information (primary triage). A Japanese OMFR specialist and a Laotian OMF radiologist examined all radiographs independently to identify abnormalities of bone of the TMJ (secondary triage).	In primary triage, patient information was insufficient 39.3%. The Japanese and Laotian OMFR experts detected TMJ abnormalities in 99 and 72 joints, respectively, with 0.84 kappa coefficient (p < 0.01).	Flaws in panoramic radiographs Incomplete patient information Lao People's Democratic Republic lack of resources, material dentists, co-dental staff, only one dental hospital. Small sample size
13. Wongwai²⁷	Cost–utility analysis	Infant with low birth weight, premature infants	Department of Ophthalmology, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand	Telemedicine for ROP screening and diagnosis	In store-and-forward telemedicine for ROP screening, trained technicians used a wide-angle digital fundus camera to capture photographs, which are then sent digitally to expert ophthalmologists for interpretation. A positive case will be referred to remote experts for further care and treatment.	The sensitivity and specificity (100% and 97.8%) of the screening and diagnosis performances of ROP using telemedicine were higher than the current method (88.9% and 93.4%). The incremental cost-effectiveness ratio was 17,397 THB per the quality-adjusted life-year rescued and 259 THB per case of avoided blindness.	Poor quality of fundus images can be reduced by photographic training and preparation of the infants before taking photographs. Small sample size
14. Lim⁸⁰	Prospective study	Patients who were implanted with CIED	National Heart Center, Singapore	Remote monitoring (telehealth)	Patients were equipped with a patient monitor with GSM network connection or telephone. Wireless and non-wireless home monitors were employed. Data were transferred to the CareLink network securely. The internet-accessible CareLink website was used to access results. Medical technicians received data transmissions.	Remote monitoring of CIED is safe and feasible. 73.7% of 334 remote transmissions scheduled were punctual compared with 53.8% of non-wireless transmissions. 73.2% of patients chose remote monitoring. Clinicians accepted that remote transmissions provided information equivalent to those conducted in-office.	Younger participants may be less technology-averse and willing to adopt remote monitoring.
15. Nguyen⁸¹	Cost-effectiveness analysis	Patients aged 55 years with type 2 diabetes previously not screened for DR	SiDRP, Singapore	Telemedicine for DR screening	SiDRP provided real-time assessment of DR photographs by expert graders. Retinal photographs were captured in primary care settings and transmitted to ocular imaging center via secured web-based teleophthalmology platform. Expert graders evaluated the severity of DR and sent results back to family physician in primary care setting.	Costs and effects of the telemedicine-based DR screening model were significantly lower (total cost savings of $173 per person); cost savings are $144 per person in direct medical cost, while the model generates equal quality-adjusted life-years.	Ungradable images Cost of time spent waiting for the result Patients do not want to wait since the results are not available on the same day.
16. Chinprasatsak⁸²	A pilot study	Patients aged between 18 and 80 years who received ALS response from Maharat Nakhon Ratchasima Hospital teams at the scenes of emergency locations before transferring to ED	Maharat Nakhon Ratchasima Hospital, a tertiary care hospital in Nakhon Ratchasima, Thailand	Telegraphic medicine	In the monitoring room at the Department of Emergency Telegraphic Medicine, the system received multiple types of data (real-time video and audio, blood pressure, pulse rate, oxygen saturation, capnography, and electrocardiography) directly from the ALS team. However, an emergency physician was not available on-site at the emergency location. An on-duty emergency clinician who was not a member of the investigation team was assigned to control the monitoring room, to communicate, and to make treatment decisions in real time with the ALS team at the scene of the emergency where the patient was located or lived.	Telegraphic medicine system significantly increased the percentage of primary diagnoses at 94.4% compared with the control group with 2.8%. The telegraphic system also increased the quality of the airway, circulation, and neurological management systems.	All data must be stored securely with a limited number of users for privacy and data security. Small sample size
17. Ng⁸³	Pre–post, quasi-interventional study	Full-time primary care nurses employed in the state of Penang, Malaysia	Primary care clinics in Penang, Malaysia	VBCI	Participants filled out the study questionnaire about the attitudes of nurses to people presenting with mental illness before and after receiving VBCI as an anti-stigma intervention via YouTube.	VBCI significantly improved the attitude of participants toward people with mental illness.	Hawthorne effect may be present. The VBCI only provided partial faithfulness to the main suggestions of an effective anti-stigma intervention. There was no measurement of sustainability, while booster VBCI would be necessary to maintain the effect.
18. Tian²⁸	Prospective study	Patients with skin problems	Singapore	Teledermatology	A store-and-forward technology was used in the system. Patients filled out questionnaire and submitted photographs of themselves through the system. An independent doctor examined patients' medical information and make diagnoses and recommendations. Another doctor examined the patients face-to-face. Diagnoses from both doctors were compared.	A store-and-forward technology of telemedicine enables dermatologists to make accurate skin diagnoses effectively compared with face-to-face diagnoses.	Poor photographic quality Physician was not fully experienced in teleconsultation.
19. Sidek⁴⁵	Cross-sectional study	Patients with mild head injury	Five peripheral hospitals without neurosurgical services in the state of Johor, Malaysia	Teleneurosurgery	Patients were admitted by the general surgeon in the primary hospital. The decision for an immediate transfer of the patients to the central NSU or to be treated in the primary hospital was made after a phone consultation between both clinicians. Clinical information and CT images were performed by teleconsultation.	11 of 359 (3.06%) patients were dissatisfied. There was no significant difference in the Glasgow Outcome Scale at 3 and 6 months for patients in the unfavorable group (p = 0.368).	Small sample size A delay in diagnosing patients
20. Sudas⁴⁶	Randomized controlled trial	Patients aged ≥20 years who were prescribed warfarin when discharged	Kamphaeng Phet Hospital, a general hospital in Thailand	Telemonitoring	Participants in the intervention group received standard pharmacy care with monthly telephone follow-up within 3 months of the study period. During each telephone call (10–25 min), clinical pharmacists asked patients about issues with administering warfarin, including adverse events, complications, and medication adherence. Pharmacists resolved patients' problems during the telephone call and provided reminders for the next scheduled visit. The control group only received standard pharmacy care.	Telephone follow-up service when combined with usual care in recently discharged patients on warfarin helped patients achieve and maintain their INR target higher than usual care only (45.6% vs. 24.1%). Telephone follow-up alone without home visits was effective in helping patients achieve and maintain their INR target.	Short period telephone follow-up Small sample size
21. Asano²⁹	A parallel, two-arm, evaluator-blinded, randomized controlled trial	Adults aged ≥40 years who had suffered a stroke within 4 weeks of the start of the study	National University of Singapore	Telerehabilitation	The telerehabilitation group received a telerehabilitation system and standardized rehabilitation that was programmed for 3 months. The standardized program included both physiotherapy and occupational therapy. An assigned therapist determined the difficulty level and minimum range of motion desired for each patient's specific exercise based on individual need.	Between the intervention and control groups, the mean differences of the Late-Life Function and Disability Instrument frequency and limitation total scores at 3 months were −3.30 (95% CI) −7.81 to 1.21) and −6.90 (95% CI −15.02 to 1.22), respectively.	The Hawthorne effect The validity and reliability of self-reported date might affect the study findings.
22. Chutinet³⁷	Descriptive study	Acute ischemic stroke patients in Lao People's Democratic Republic	Mittaphab Hospital, Lao People's Democratic Republic KCMH, Thailand	Telestroke	Teleconsulting team at KCMH was available 24 h a day/7 days a week. On-site training at Mittaphab hospital was performed to ensure understanding of acute ischemic stroke, stroke fast track system, thrombolytic treatment, and post-thrombolytic care to physicians, nurses, and health care workers. The video telecommunication used the LINE application for real-time communications.	10 patients were eligible for intravenous rt-PA. The average of the NIHSS before thrombolysis was 10. The mean DTN time was 108 min, and the mean duration from onset to hospital arrival was 122.5 min.	Small sample size The awareness of and access to telestroke are still lacking in the country, which should be endorsed by policy makers.
23. Mong³⁸	Observational study	Patients aged ≥21 years who presented with acute ischemic stroke	CGH, a 1,000-bed acute regional hospital in the eastern part of Singapore	Telestroke	In the hub-and-spoke model, where eligible patients presenting to CGH (i.e., spoke) with acute ischemic stroke are evaluated by off-site neurologists from the NNI, Singapore (i.e., hub). The web-based telemedicine system is used for real-time communication and remote image transmission.	The median DTN time improved from 96 to 78 min post-implementation (p = 0.003). Fewer patients had symptomatic intracranial hemorrhages and more patients had improvements in their National Institutes of Health Stroke Scale score.	Health care staff should become more familiar with the practice during the study time. Frequency of software or equipment failure resulted in delays. Small sample size
24. Ferraris³⁰	Cross-sectional study	Telemedicine beneficiaries	Jose R. Reyes Memorial Medical Center, Manila, Philippines	Teleconsultation	Neurosurgical residents and neurosurgical patients received communications via free online instant messengers to obtain outpatient, symptom assessment, functions of triaging, follow-up care, medication adjustment, and other functions. Telehealth providers learned about telemedicine via online modules and training materials.	The answers to survey questions about neurosurgical outpatient care throughout telemedicine returned high ratings of effectiveness for the patients and their caregivers. The affordability of the telemedicine setup saved the patients from borrowing money and avoiding adverse events.	N/A
25. Jang³⁹	Cost–benefit analysis	Korean physicians worked in Vietnam	Hospital in Korea and Hospital in Vietnam	Robot-based telemedicine consultation service	Korean physicians only stay in Vietnam for a few days to conduct surgery. They return to Korea without providing pre- and postoperative care. Robot-based telemedicine consultation services would allow Korean physicians to provide pre- and postoperative medical care without physically visiting Vietnam, thus enabling them to provide these services without additional travel costs.	The annual profit from replacing conventional outpatient services was predicted as 1,508.51 USD, and the annual budget of a robot-based telemedicine consultation system was predicted at 6,056.40 USD.	Small sample size Qualitative analyses are needed such as patient outcomes, quality of care, and patient satisfaction, which would provide a more complete evaluation of the service.
26. Lim⁴⁰	Descriptive study	Adolescents with diabetes mellitus	Two public tertiary pediatric hospitals in Singapore	Nurse-led telehealth	The important domains of the telehealth service during the pandemic included (1) teleconsultation platform, (2) payment platform, (3) laboratory tests, (4) medication collection/delivery. The system began with an APN sending a text message to inform about the clinics conversion to telehealth service. After verbal consent was obtained, the APN made a request for patients to submit home blood sugar results electronically if available. The patients were required to come to the hospital only for laboratory tests and to collect medications. Prescriptions were prefilled and transmitted electronically via a closed-loop prescribing system setup with the hospital pharmacy or home delivery service. Once HbA1c results were available, the APN made an audio call back to patients to provide clinical information.	The adolescents felt very satisfied with the administrative processes of the telehealth service and spending less time in the hospital greatly reduced their anxiety around contracting COVID-19. There were no emergency visits for diabetes-related adverse events, and HbA1c levels remained stable.	There were uncontactable patients.
27. Lin⁴¹	Descriptive study	COVID-19	UCC at Alexandra Hospital, Singapore	Telemedicine (teleconferencing)	The use of telemedicine via videoconferencing to facilitate history taking for patients in the ESA. Health care workers communicated with patient between the main building and ESA using teleconferencing with internet-capable laptops. A physician entered the ESA to examine the patient, and a nasopharyngeal swab was taken if necessary.	The amount of time for the ESA was reduced by using telemedicine. It also helped to reduce utilization of PPE resources, which may were in short supply during the pandemic.	N/A
28. Low²⁵	Descriptive study	Geriatric patients	Nursing homes and the geriatric department of KTPH, a 590-bed acute general hospital, North of Singapore, Singapore	Telegeriatric	The telemedicine system used a hub (KTPH) and spoke (nursing homes) model. Teleconsultation sessions were performed with each nursing home two to three times a month. The telegeriatric nurse presented the patient's clinical information to the specialists. Doctor and nurse discussed and completed diagnoses and a care plan on the TCF. Then, the doctor reviewed and signed off on the TCF and send it back to the nurse via email.	Telemedicine services for older patients are reasonable. The services provide high quality for home care, nursing, and rehabilitation centers.	Self-reported may be misreporting. Documentation within the TCF was inconsistent and incomplete between different doctors. An electronic version of the TCF along with a dropdown list of options could be applied in the future to standardize terms and terminologies.
29. Mappangara⁴²	Cross-sectional study	Patients with a cardiovascular symptom	Makassar Cardiac Center in collaboration with the local government of Makassar City provided one digital ECG machine for each primary care center in Indonesia	Tele-ECG consultation	The ECGs were sent from the primary centers to the cardiac center daily. Patients were referred if they had a cardiovascular disease symptom, and the ECG showed an abnormal finding. The advice was to refer the patient to a hospital with cardiovascular care facilities when the patient was uncertain about the ECG. The cardiologist took responsibility for the diagnosis and treatment after patients were referred from the primary care hospital.	Tele-ECG consultation was beneficial to support primary care providers in making a quick pre-hospital triage among Indonesian patients with low-to middle income. Tele-ECG is associated with a higher rate of early hospitalization when required.	There was unorganized and incomplete patient data from primary care centers. Primary care centers with no electronic health records available and lack of standardized and reliable medical records. The primary care level should have the detection and screening of diabetes and dyslipidemia available.
30. Quek⁴³	Retrospective study	Patients requiring ⁹⁰Y selective internal radiation therapy	Singapore	Teleradiology	NM physicians outside the institution are not allowed to be on-site. The case history and images were electronically transmitted using a secured encrypted channel from the institution to the off-site NM physician. NM expertise is visually present using videoconferencing to verify the configuration of the delivery system and the appropriate sites of delivery of ⁹⁰Y and to monitor radioactive therapy by interventional radiologists.	Use of telemedicine allowed the institution to continue collaborating with the off-site NM physician partners in the delivery of oncological care to patients during the COVID-19 outbreak.	N/A
31. Singh³¹	Retrospective study	HD patients	Community-based HD centers, Singapore	Tele-HD	Tele-HD, virtual patient rounds using tablet computers were conducted via videoconferencing with the nephrologist while patients and dialysis nursing staff remained physically at the dialysis centers during COVID-19. The results of laboratory and dialysis studies were evaluated on an electronic medical record with logged progress notes.	The hospitalization rate was 13.9% (95% CI: 5.6–21.5, p < 0.001) lower in the period after telemedicine rounds were introduced. There was no significant increase in mortality during the months when telemedicine was performed.	A short duration of follow-up. Some potential errors in categorizing patients, such as reasons for hospitalizations and the exact cause of death, were inevitable as data were extracted from electronic medical records and related data systems.
32. Tan³²	Descriptive study	Older adults	Singapore	Telehealth consultations	Strategies to enhance greater uptake of telehealth consultations among older adults during COVID-19 pandemic: Collaborations with nearby community clinics A smooth and seamless workflow Medication delivery to patients' homes Telehealth services included a variety of health care services.	The COVID-19 pandemic led to the increased uptake of telehealth services among older adults due to safety, convenience, and time and travel savings. The greater uptake increased staff familiarity with the workflows. Increased utilization of telehealth allowed institutions to devote more resources to developing telehealth services.	Clinical examination and assessment are the limitations of telehealth. Not all older adults have blood pressure gauges at home. Longitudinal measurement of vital signs is difficult to perform by telehealth.
33. Tan³³	Descriptive study	Patients who were seen by Tele-SOC over a 2-week period	Geriatric Medicine clinic in an acute hospital, Singapore	Tele-SOC	AES encryption was operated for Tele-SOC consultations. At the hospital's Tele-SOC unit, two laptops with Intel Core i5 processors were used with the laptops' webcam, speaker, and microphone. Patients and/or caregivers were instructed to download the videoconferencing application to their smartphones or laptops with the help of written instructions. The Tele-SOC center was conducted by one geriatrician and assisted by a geriatric nurse, who were trained to set up the system and to troubleshoot technical problems associated with the videoconference application.	Tele-SOC provided a good alternative to physical consultations for the geriatric population to minimize their contact with viral transmissions. Communication was satisfactory for most patients and caregivers.	A lack of physical examination and accurate clinical findings A small sample size
34. Xie⁸⁴	Economic analysis	Consecutive patients with diabetes	Singapore	Teleophthalmology-based DR screening	SiDRP is a national DR screening program, which was established to provide screening for patients within the network of primary care clinics. SiDRP uses a teleophthalmology platform. A nurse practitioner uses a retinal camera to capture retinal images from undilated pupils. The digital retinal images are transferred from the primary eye care clinic to a centralized eye center for human evaluation of the occurrence of DR. Deep learning systems were performed to identify DR issues that could be referred to the eye center including the semi-automated deep learning system model (human and deep learning screening) and the fully automated model (only deep learning screening).	The semi-automated screening model was the least expensive of the three models, at US $62 per patient per year. The fully automated model was $66 per patient per year, and the human assessment model was $77 per patient per year. Human assessment system, the semi-automated system, and fully automated system had equal sensitivity 89.9 (95% CI: 85.9–93.1), whereas they had different specificity 99.2 (95% CI: 99.0–99.3), 99.6 (95% CI: 99.4–99.7, 81.8 (95% CI: 81.1–82.4), respectively)	Referral bias
35. Lam³⁴	Descriptive study	General patients	Endocrinology Department, Singapore	Video consultation	Videoconsultations were used, as this afforded physicians the ability to read visual cues and perform visual assessments. Zoom was used to conduct videoconsultations. Physicians reviewed all upcoming outpatient appointments and decided whether to proceed with original in-person appointments, or to defer nonurgent appointments.	Verbal feedback from patients reflected general satisfaction with the convenience afforded by videoconsultations. They saved on transportation costs and time spent.	Disabling both meeting recordings and in meeting file transfer features; and allowing the meeting host to control content sharing. Videoconsultations required more time per patient for physicians and PSAs than in-person consultations.
36. Lian³⁵	Descriptive study	Diabetes patients	Tan Tock Seng Hospital, Singapore	Telehealth	Tan Tock Seng Hospital's telehealth platform. The virtual health system included remote monitoring, consultations by telephone, and video telehealth visits using a smartphone.	Virtual health applications were found to be safe, effective, and efficient for replacing in-person visits.	N/A
37. Surapat⁴⁷	Commentary/descriptive study	COVID-19 patients	A tertiary care hospital in Thailand	Telemonitoring	Pharmacists conducted regular telemonitoring for each patient via the hospital's CPOE system, and then, they accessed information related to the patient's condition. Pharmacists provided suggestions by telephone call directly to physicians or to the ward. The clinical pharmacists, physicians, and nurses communicated about cases and medications by using online messenger or phone call.	Pharmaceutical services can be provided to COVID-19 patients by clinical pharmacists using telemonitoring. The technology can reduce direct patient contact.	Clinicians working at wards should be urged to enter information as much as possible on patients' progression. This information would facilitate patient care services by multidisciplinary teams. Patients' privacy and authority to access patient information must be protected thoroughly.

AES, Advanced Encryption Standard; ALS, advanced life support; APN, advanced practice nurse; CGH, Changi General Hospital; CI, confidence interval; CIED, cardiac implantable electronic devices; CMV, cytomegalovirus; COVID-19, coronavirus disease 2019; CPOE, computerized physician order entry; CSC, Children's Surgical Centre; CT, computed tomography; DFP, digital fundus photograph; DTD, door to decision; DTN, door to needle; ECG, electrocardiogram; ED, emergency department; ESA, extended screening area; GSM, Global System for a Mobile communications; HbA1c, hemoglobin A1C; HD, hemodialysis; INR, international normalized ratio; iPATH, Internet Pathology Suite; KCMH, King Chulalongkorn Memorial Hospital; KTPH, Khoo Teck Puat Hospital; LED, light-emitting diode; MMS, multimedia messaging system; MRI, magnetic resonance imaging; N/A, insufficient data or no information; NIHSS, National Institute of Health Stroke Scale; NM, nuclear medicine; NNI, National Neuroscience Institute; NSU, neurosurgical unit; OMFR, oral and maxillofacial radiology; PACS, picture archiving and communication system; PPE, personal protective equipment; PR, panoramic radiography; PSAs, Patient Service Associates; ROP, retinopathy of prematurity; rt-PA, recombinant tissue-plasminogen activator; SiDRP, Singapore Integrated Diabetic Retinopathy Programmed; SMS, short message service; TCF, telegeriatric consultation form; Tele-SOC, Telemedicine Specialist Outpatient Clinic; THB, Thai Baht; TMJ, temporomandibular joint; TTSH, Tan Tock Hospital; UCC, urgent care center; UHS, University of Health Sciences; VBCI, video-based contact intervention.

TELEMEDICINE MODEL

A hub-and-spoke system was one of the models applied to developing and using telemedicine systems. Ang et al. used the hub-and-spoke model for telestroke in Singapore. An acute stroke patient at the emergency department (ED) of a study hospital (spoke) was examined by the duty neurologist at the National Neuroscience Institute (hub) via a high-definition webcam.²⁴ The neurologist viewed the image of the patient's computed tomography brain scan via high-speed internet connection. An ED physician assisted and communicated with the neurologist in real time until a diagnosis of thrombolysis was reached.

Low et al. also applied this model to the telegeriatric system in Singapore. The telemedicine system used a hub (hospital) and spoke (nursing home) model.²⁵ Teleconsultation sessions were performed with each nursing home two to three times per month. The telegeriatric nurse presented the patient's clinical findings to the specialists, and the doctor and nurse completed the diagnostic workup and care plan on the telegeriatric consultation form. Then, the doctor reviewed and signed off on the telegeriatric consultation form and sent the form back to the nurse via email.

A store-and-forward model was also used for telepathology in Cambodia,²⁶ telemedicine for retinopathy in Thailand,²⁷ and teledermatology in Singapore.²⁸ Kadaba et al. used a hybrid telepathology network based on store-and-forward communication using the Internet Pathology Suite (iPATH) network and videoconferencing using Skype (Microsoft Corporation, Luxembourg). Tissue slides were prepared at a Cambodian laboratory, and images were captured and uploaded to the iPATH network along with diagnostic comments. Specialists worldwide who were invited to join the iPATH network were notified of the meeting and used the opportunity to provide a second opinion on the cases. In addition, telemonitoring was used to monitor patients who were implanted with cardiac implantable electronic devices.

COMMUNICATION

Included studies used several methods for remote communication between patients and doctors, or between health care providers. Video consultation was used in many telemedicine systems for communication between patients and health care providers.^{24,28

–35} Videoconferencing or teleconferencing was used to communicate between health care providers, internal and external to the organization, to discuss with patients.^{25,26,31,33

–43} Telephone and SMS were the most common systems used by health care providers to contact patients and other health care partners.^{35,36,40,44

–47}

TECHNOLOGY

Skype and Zoom (Zoom Video Communications, Inc.) were used for videoconferencing. Line (LINE Corporation, Japan) was used for real-time communication in both chat and video call. Email and SMS were used to communicate between nursing homes and physicians. YouTube (Google LLC) was used to provide education for health care staff and patients. A digital fundus camera was the main device used in teleophthalmology. Some studies developed their own communication methods for use with telemedicine systems. For example, Tan et al. developed the teleophthalmology software that allowed offline data retrieval and real-time sampling of patient data.³⁶ The data were transferred from the primary care centers to the main hospital securely via hospital networks in Singapore.

ISSUES AND LIMITATIONS

Most included studies had a small sample size and were of short duration. The Hawthorne effect, which is a change in behavior as a motivational response to the interest, care, or attention received through observation, assessment, or experiment, may be present.⁴⁸ Implementation of complete telestroke with both video-teleconferencing and teleradiology is expensive for developing countries to operate and sustain. Low-quality images or unreadable images diminished the ability of physicians to effectively evaluate patients' presentations. In addition, telemedicine software systems and equipment were frequently down resulting in delays in service and diagnoses. Staff members were not familiar with the use of telemedicine equipment, and physicians were not experienced in using teleconsultation programs. Each of these elements diminished the potential for telemedicine having a strong positive impact on SEA patients.

The ability to effectively conduct a clinical examination and assessment is the major limitation in telehealth, resulting in inaccurate physical examinations and clinical findings. Clinical staff often did not have time to obtain informed consent, and as a result, the documentation produced was inconsistent and incomplete. Moreover, primary care centers did not have electronic health records available and lacked standardized, reliable medical records. Patients were easily lost to follow-up or difficult to track when they were transferred to another hospital, and some patients were unable to understand the recommendations that health professionals provided by telephone. Technological barriers, inability to use technology, and difficulty accessing a good quality internet connection were additional common problems.

QUALITY OF INCLUDED STUDIES

Utilizing the STARE-HI principles,²² two risks of bias graphs were created and summarized (Figs. 3 and 4). Almost half of the studies were at high risk or unclear due to the methods used to perform the data analysis and the study design, whereas most of the studies were at low risk of bias in a participant domain. For study design, 24 studies (64%) were at low risk of bias and 4 studies (11%) were at high risk. For theory and background, 25 studies (68%) were at low risk of bias and 12 studies (32%) were at high risk. We identified several studies (87%) that adequately described the method for selecting participants, users, patients, units, or hospitals.

Fig. 3.

Risk of bias graph: review authors' judgment about each domain presented as percentage across included studies.

Fig. 4.

Risk of bias graph: review authors' judgment about each domain for each included study.

Twenty-nine studies (78%) provided sufficient details on study flow, including study period, study activities, timeline of activities, or flow diagram. Twenty-six studies (70%) clearly defined the outcome or other variable measures that were used in the studies, whereas 10 studies (27%) were at high risk. Methods for data acquisition and measurement, such as questionnaires, interviews, observations, type of data collection, were described sufficiently in 27 studies (73%).

For methods of data analysis, 14 studies (38%) were at high risk of bias as they did not describe details of data analysis in terms of quantitative and qualitative methods. Furthermore, we found that many studies were at high risk of bias because of small sample sizes.

Discussion

As a result of the COVID-19 pandemic, the authors observed that the number of research studies conducted on the various aspects of telemedicine increased in 2020. Overall, most of the included studies were conducted in Singapore (49%). This systematic review also demonstrated that a hub-and-spoke modality was widely used throughout the SEA region, and the majority of telemedicine systems were hospital-based with teleophthalmology being the most common, followed by teleconsultation. A descriptive study design predominated, followed by retrospective and cross-sectional studies.

The World Health Organization ranked the national health care system in Singapore as the sixth best system in the world based on the program's excellent medical ICT capabilities.^49,50 Among Asian countries, Singapore is a leader in applying information technology to the health care industry. In 2000, the government deployed the Singapore One Infrastructure, which built and enabled an e-government and e-health environment^49,51 that included high-speed broadband internet and appropriate technology to implement telemedicine applications throughout the country.⁵²

The National Telemedicine Guidelines (NTG) addressed the relevant components for the delivery of telemedicine services by health care providers through four domains: “Clinical Standards and Outcomes,” “Human Resources,” “Organisational,” and “Technology and Equipment.” The NTG is the only guideline among SEA countries that addresses the scalability, maintenance of the technology, equipment calibration, end of life and e-waste disposal, upgrade and replacement of appropriate technology, and equipment.^18,52

The hub-and-spoke model was the most widely used telemedicine system.^53,54 Although the model provides many benefits, it has risks due to congestion at hubs, overextension of spokes, staff dissatisfaction at spokes, and transportation disruptions.⁵³ The store-and-forward concept, in which medical information is collected, compiled, and transferred through the internet, is also used in many telemedicine systems.⁵⁵ This concept is the main methodology for deploying telemedicine services in Singapore.⁵⁶ Moreover, Deldar's systematic review of teleconsultation and clinical decision-making revealed that the store-and-forward model was used by 31% of included studies.⁵⁷

Ibrahim et al. reported that most health care professionals were dissatisfied with the reduction in face-to-face communication with their patients.⁵⁸ Ibrahim et al. also noted that the success of telemedicine depended on the perception and knowledge of the participating health care professionals.^58,59 Our study found that staff members and/or physicians were uncomfortable using telemedicine or teleconsultation equipment²⁸ probably because they had limited familiarity with the technology. Thus, the lack of comfort and skills of health care workers clearly had a negative effect on the adoption of telemedicine systems and services in SEA countries.⁶⁰

Similarly, a systematic review by Durrani and Khoja of telehealth in Asian countries found that untrained service providers, including physicians and staff, were the most significant barriers to the successful implementation of telehealth.⁶¹ Health care organizations and institutions should provide training, instructional materials, and resources about telemedicine to ensure that staff are capable of using the system effectively. Telemedicine has the potential to have a greater impact on health care practice in developing countries than in developed countries⁵⁹ if health care professionals could be motivated to learn how to use the systems effectively in both rural and urban areas.⁵⁸

Telemedicine programs in developing countries were frequently small-scale pilot or informal projects with limited implementation.¹² High telecommunication and device costs were the major barriers to implementing telemedicine systems in many Asian countries,⁶¹ and ICT resources were frequently unavailable in clinical units.⁶² By contrast, health care providers and patients frequently used a telephone, SMS, multimedia messaging system, Line, Facebook Messenger, iMessage, WhatsApp, Chatbot, and email to communicate with each other.^37,62
–64

Health care providers also used messaging applications in their daily work to share information with colleagues.⁶⁵ Although using free messaging applications provided many advantages to improving communication, it likely increased the risk of exposing data to privacy and security abuses.^65
–67 While the applications available are free, they may not offer the sophisticated technology that is needed to conduct effective telemedicine patient–physician interactions in a safe and secure environment.⁶⁸

The inability to conduct clinical and physical examinations was the most serious limitation of telemedicine when compared with face-to-face interactions. For example, doctors and therapists experienced difficulty evaluating and managing several cases using telerehabilitation because of their inability to operate the program. Both patients and providers lacked the knowledge, skill, and technical know-how to interact during telemedicine services.⁶⁹

Another significant barrier perceived by the therapists was patient assessments, such as the inability to physically test muscle power or balance⁷⁰ as well as the barriers of interaction between the patient and the therapist. A mixed method study conducted by Lee et al.⁷¹ found that most participants preferred telemonitoring.

Clearly, technology issues are the biggest barriers to successful implementation of telemedicine. Finally, it is important to note that some participants were concerned about sharing their data with family members and were concerned about malware that could infiltrate telemonitoring devices and attack their personal information.

Conclusions

The review was conducted using a rigorous methodology and transparency. A comprehensive literature search was conducted using a variety of databases and related sources to retrieve all related studies on telemedicine in SEA countries. The present study had some limitations. First, we did not include studies of m-health or e-health as the goal was to focus on the overall implementation of telemedicine systems. Second, articles were included only from the last 11 years of published peer-reviewed studies. Third, no gray literature was included. Moreover, a meta-analysis could not be conducted because of heterogeneity among included studies.

The overall finding of the present review provides an important perspective on the use and implementation of telemedicine across SEA settings. Studies on clinical applications of telemedicine are critical to inform the usability and quality of the systems, but there was limited information and a lack of good quality studies to assess. Most research efforts were conducted with a small sample size without calculation, which had no power to estimate the difference. Thus, generalizability is limited to the study settings only.

The review illustrates that telemedicine is a powerful tool. It is feasible, safe, effective, and less expensive than traditional methods. However, the unintended consequences of health information technology⁷² are less likely to be reported in telemedicine studies. Robust research is needed to fully investigate more effective training, patient safety, use, and implementation of telemedicine systems in the Southeast Asia region.

Footnotes

Acknowledgments

We gratefully acknowledge Ross Koppel, PhD, for his valuable comments and suggestions to our project. We would like to thank Dr. Monique Saran for her idea to explore the topic of telemedicine.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

WHO Global Observatory for eHealth. Telemedicine: opportunities and developments in Member States: Report on the second global survey on eHealth. Geneva: World Health Organization, 2010.

Sood

, Mbarika

, Jugoo

, et al. What is telemedicine? A collection of 104 peer-reviewed perspectives and theoretical underpinnings. Telemed J E Health, 2007; 13:573–590.

The Center for Connected Health Policy. State telehealth laws and reimbursement policies. In: Kwong MW, ed. West Sacramento, CA: Public Health Institute, 2020.

de la Torre-Coronado

, Vaca

, et al. Cost-utility and cost-effectiveness studies of telemedicine, electronic, and mobile health systems in the literature: A systematic review. Telemed J E Health, 2015; 21:81–85.

Strehle

, Shabde

. One hundred years of telemedicine: Does this new technology have a place in paediatrics?. Arch Dis Child, 2006; 91:956–959.

Robinson

, Seale

, Tiernan

, Berg

. Use of telemedicine to follow special needs children. Telemed J E Health, 2003; 9:57–61.

Mishra

. Telemedicine in otolaryngology (An Indian Perspective). Indian J Otolaryngol Head Neck Surg, 2003; 55:211–212.

Brandling-Bennett

, Kedar

, Pallin

, et al. Delivering health care in rural Cambodia via store-and-forward telemedicine: A pilot study. Telemed J E Health, 2005; 11:56–62.

ACCESS Health International Southeast Asia. Reaching 650 million: How digital technology is key to achieving universal health coverage in ASEAN. Singapore: Cisco Systems, 2019.

10.

Witoolkollachit

. eHealth strategy, ministry of public health (2017– 2026). In: Witoolkollachit

, ed. Vol. 1. Nonthaburi. Thailand: Information and Communication Technology Center, Office of the Permanent Secretary, Ministry of Public Health, 2017.

11.

Koh

. Thailand launches telemedicine programme in rural areas. 2021. Available at https://www.healthcareitnews.com/news/apac/thailand-launches-telemedicine-programme-rural-areas (last accessed September 12, 2021 ).

12.

Nguyen

, Vu

, Dao

, et al. Digital immunization registry: Evidence for the impact of mHealth on enhancing the immunization system and improving immunization coverage for children under one year old in Vietnam. mHealth, 2017; 3:26.

13.

Kijsanayotin

, Kasitipradith

, Pannarunothai

. eHealth in Thailand: The current status. Stud Health Technol Inform , 2010; 160:376–380.

14.

Scott

, Mars

. Telehealth in the developing world: Current status and future prospects. Smart Homecare Technol Telehealth, 2015; 3:25–37.

15.

Suhaiza Zailani

, Omar

. Telemedicine in the Malaysian public hospitals: Are we there yet? In: International Conference on Business, Finance and Tourism Management. Vol. 2009: WASET, 2009.

16.

Sheng

ORL

, Hu

PJH

, Chih-Ping

, Pai-Chun

. Organizational management of telemedicine technology: Conquering time and space boundaries in health care services. IEEE Trans Eng Manag, 1999; 46:265–278.

17.

Smith

, Thomas

, Snoswell

, et al. Telehealth for global emergencies: Implications for coronavirus disease 2019 (COVID-19). J Telemed Telecare, 2020; 26:309–313.

18.

Intan

, Defi

. Telemedicine guidelines in South East Asia—A scoping review. Front Neurol, 2020; 11:581649.

19.

Romanick-Schmiedl

, Raghu

. Telemedicine—Maintaining quality during times of transition. Nat Rev Dis Primers, 2020; 6:45.

20.

Analytics

. Endnote, X9 ed. Philadelphia, USA: Clarivate Analytics, 2019.

21.

Higgins

JPT

, Thomas

. Cochrane handbook for systematic reviews of interventions, 2, Chichester (UK): John Wiley & Sons, 2019.

22.

Talmon

, Ammenwerth

, Brender

, et al. STARE-HI—Statement on reporting of evaluation studies in Health Informatics. Int J Med Inform, 2009; 78:1–9.

23.

Higgins

JPT

, Savović

, Page

, et al. Assessing risk of bias in a randomized trial. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, eds. Cochrane handbook for systematic reviews of interventions, Chichester (UK): John Wiley & Sons, 2019:205–228.

24.

Ang

, Tan

, Singh

. Telestroke: Rapid treatment of acute ischemic stroke patients using telemedicine in a Singapore emergency department. Eur J Emerg Med, 2013; 20:322–326.

25.

Low

, Toh

, Tan

LLC

, et al. The nuts and bolts of utilizing telemedicine in nursing homes—The GeriCare@North experience. J Am Med Dir Assoc, 2020; 21:1073–1078.

26.

Kadaba

, Ly

, Noor

, et al. A hybrid approach to telepathology in Cambodia. J Telemed Telecare, 2013; 19:475–478.

27.

Wongwai

, Kingkaew

, Asawaphureekorn

, Kolatat

. A store-and-forward telemedicine for retinopathy of prematurity screen: Is it cost-effective in Thailand?. Asian Biomed, 2015; 9:665–673.

28.

Tian

. Tele-aesthetics in South Asia. J Cosmet Dermatol, 2017; 16:21–25.

29.

Asano

, Tai

, Yeo

, et al. Home-based tele-rehabilitation presents comparable positive impact on self-reported functional outcomes as usual care: The Singapore Tele-technology Aided Rehabilitation in Stroke randomised trial. J Telemed Telecare, 2019; 27:231–238.

30.

Ferraris

, Golidtum

, Zuñiga

BKW

, et al. Recapitulating the Bayesian framework for neurosurgical outpatient care and a cost-benefit analysis of telemedicine for socioeconomically disadvantaged patients in the Philippines during the pandemic. Neurosurg Focus, 2020; 49:E14.

31.

Singh

, Ngoh

, Wong

, Khan

. Impact of telemedicine on hospitalisation and mortality rates in community-based haemodialysis centres in Singapore during the COVID-19 pandemic. Ann Acad Med Singap, 2020; 49:756–763.

32.

Tan

, Ho Wen Teng

, Seetharaman

, Yip

. Facilitating telehealth for older adults during the COVID-19 pandemic and beyond: Strategies from a Singapore geriatric center. Geriatr Gerontol Int, 2020; 20:993–995.

33.

Tan

, Pillay

, Chia

, et al. Using telemedicine for outpatient geriatric care during the novel coronavirus outbreak: Experience from the first 15 patients. Ann Acad Med Singap, 2020; 49:699–703.

34.

Lam

AYR

, Chan

ECL

, Quek

CMX

, et al. Videoconsultation to overcome barriers during COVID-19. Ann Acad Med Singap, 2021; 50:77–83.

35.

Lian

, Dalan

, Seow

, et al. Diabetes care during COVID-19 pandemic in Singapore using a telehealth strategy. Horm Metab Res, 2021; 53:191–196.

36.

Tan

JCH

, Poh

EWT

, Srinivasan

, Lim

. A pilot trial of tele-ophthalmology for diagnosis of chronic blurred vision. J Telemed Telecare, 2013; 19:65–69.

37.

Chutinet

, Keosodsay

, Vorasayan

, et al. The first 10 thrombolysis for acute ischemic stroke in Lao People's Democratic Republic under teleconsultation from Thailand. J Stroke Cerebrovasc Dis, 2019; 28:104327.

38.

Mong

, Tiah

, Wong

, Tan

. Improving telestroke treatment times through a quality improvement initiative in a Singapore emergency department. Singapore Med J, 2019; 60:69–74.

39.

Jang

, Lee

, Hong

, et al. Economic evaluation of robot-based telemedicine consultation services. Telemed J E Health, 2020; 26:1134–1140.

40.

Lim

, Yap

, Chin

. Bridging the needs of adolescent diabetes care during COVID-19: A nurse-led telehealth initiative. J Adolesc Health, 2020; 67:615–617.

41.

Lin

, Sim

, Ong

VYK

, et al. Telemedicine in the acute care setting during the COVID-19 pandemic. Intern Emerg Med, 2020; 15:1591–1593.

42.

Mappangara

, Qanitha

, Uiterwaal

CSPM

, et al. Tele-ECG consulting and outcomes on primary care patients in a low-to-middle income population: The first experience from Makassar telemedicine program, Indonesia. BMC Fam Pract, 2020; 21:247.

43.

Quek

, Kannivelu

, Pua

. 90Y radioembolization: Telemedicine during COVID-19 outbreak, opportunity for prime time. J Nucl Med, 2020; 61:780.

44.

Wai Leng

, Jundong

, Li Wei

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

45.

Sidek

MSM

, Siregar

, Ghani

ARI

, Idris

. Teleneurosurgery: Outcome of mild head injury patients managed in non-neurosurgical centre in the State of Johor. Malays J Med Sci, 2018; 25:95–104.

46.

Sudas Na Ayutthaya

, Sakunrak

, Dhippayom

. Clinical outcomes of telemonitoring for patients on warfarin after discharge from hospital. Int J Telemed Appl, 2018; 2018:7503421.

47.

Surapat

, Sungkanuparph

, Kirdlarp

, et al. Role of clinical pharmacists in telemonitoring for patients with Coronavirus Disease 2019 (COVID-19). J Clin Pharm Ther, 2021; 46:236–239.

48.

Sedgwick

, Greenwood

. Understanding the Hawthorne effect. BMJ, 2015; 351:h4672.

49.

Tandon

, Murray

CJL

, Lauer

, Evans

. Measuring overall health system performance for 191 countries: GPE discussion paper series. Vol. 30. Geneva, Switzerland: WHO, 2000:23.

50.

, Jarvis-Selinger

, Lauscher

, Cordeiro

, Scott

. Technology enabled knowledge translation for eHealth: principles and practice. New York, NY: Springer Science & Business Media. Springer, 2012.

51.

Weiling Ke

KKW

. Successful e-government in Singapore. Commun ACM, 2004; 47:95–99.

52.

Nguyen

, Naguib

, Abd Ghani

, et al. An analysis of the healthcare informatics and systems in Southeast Asia: A current perspective from seven countries. Int J Electron Healthc, 2008; 4:184–207.

53.

Elrod

, Fortenberry

. The hub-and-spoke organization design: An avenue for serving patients well. BMC Health Serv Res, 2017; 17:457.

54.

Ahlquist

G SS

, Belokrinitsky

, Kapur

. Charting a clear course in rough seas: a new view on hospital and health systems strategy. New York: Strategy&, 2012.

55.

Bangert

, Doktor

. Implementing store-and-forward telemedicine: Organizational issues. Telemed J E Health, 2000; 6:355–360.

56.

Abd Ghani

, Mostafa

, Mustapha

, Aman

, Abed Mohamed

, Musa Jaber

. Investigating telemedicine approaches: A 10-country comparison. In: Ahlquist

, Saxena

, Belokrinitsky

, Kapur

, eds. Charting a clear course in rough seas: A new view on hospital and health systems strategy, Vol. 7. New York: Strategy &, 2018:224–234.

57.

Deldar

, Bahaadinbeigy

, Tara

. Teleconsultation and clinical decision making: A systematic review. Acta Inform Med, 2016; 24:286–292.

58.

Ibrahim

MIM

, Phing

, Palaian

. Evaluation of knowledge and perception of Malaysian health professionals about telemedicine. J Clin Diagn Res, 2010; 4:2052–2056.

59.

Edworthy

. Telemedicine in developing countries. BMJ, 2001; 323:524–525.

60.

Hariyati

RTS

, Sahar

. Perceptions of nursing care for cardiovascular cases, knowledge on the telehealth and telecardiology in Indonesia. Int J Collab Res Intern Med Public Health, 2012; 4:116–128.

61.

Durrani

, Khoja

. A systematic review of the use of telehealth in Asian countries. J Telemed Telecare, 2009; 15:175–181.

62.

Jantavongso

. Ethics, social media and e-health in Thailand. J Thai Med Inform Assoc, 2015; 1.

63.

Suksmono

, Sastrokusumo

, Mengko

, et al. Overview of telemedicine activities in Indonesia: progress and constraints: In; Proceedings. 6th International Workshop on Enterprise Networking and Computing in Healthcare Industry-Healthcom 2004 (IEEE Cat. No. 04EX842): IEEE, 2004:37–40.

64.

Narenthiranathan

, Adnan

, Haspani

MSM

. Tailoring teleconsultation to meet the current needs of neurosurgical services: A multimodality oriented neurosurgical consultation. Stud Health Technol Inform, 2010; 161:112–121.

65.

De Benedictis

, Lettieri

, Masella

, et al. WhatsApp in hospital? An empirical investigation of individual and organizational determinants to use. PLoS One, 2019; 14:e0209873.

66.

Dhuvad

, Dhuvad

, Kshirsagar

. Have smartphones contributed in the clinical progress of oral and maxillofacial surgery?. J Clin Diagn Res , 2015; 9:Zc22–Zc24.

67.

Bassan

. Data privacy considerations for telehealth consumers amid COVID-19. J Law Biosci, 2020; 7:lsaa075.

68.

Thimbleby

. Achieving the digital promise: Stories of digital healthcare. Birmingham, UK: Oxford University Press, 2020.

69.

Leochico

CFD

. Adoption of telerehabilitation in a developing country before and during the COVID-19 pandemic. Ann Phys Rehabil Med, 2020; 63:563–564.

70.

Tyagi

, Lim

DSY

, Ho

WHH

, et al. Acceptance of tele-rehabilitation by stroke patients: perceived barriers and facilitators. Arch Phys Med Rehabil, 2018; 99:2472–2477.e2472.

71.

Lee

, Chan

CKY

, Chua

, et al. Using telemedicine to support care for people with type 2 diabetes mellitus: A qualitative analysis of patients' perspectives. BMJ Open, 2019; 9:e026575.

72.

Harrison

, Koppel

, Bar-Lev

. Unintended consequences of information technologies in health care—An interactive sociotechnical analysis. J Am Med Inform Assoc, 2007; 14:542–549.

73.

Muengtaweepongsa

, Dharmasaroja

, Maungboon

, Wattanaruangkowit

. Feasibility and safety of remote radiology interpretation with telephone consultation for acute stroke in Thailand. Neurol India, 2010; 58:740–742.

74.

Ausayakhun

, Skalet

, Jirawison

, et al. Accuracy and reliability of telemedicine for diagnosis of cytomegalovirus retinitis. Am J Ophthalmol, 2011; 152:1053.e1–1058.e1.

75.

Shah

, Leo

, Pan

, et al. Telemedicine screening for cytomegalovirus retinitis using digital fundus photography. Telemed J E Health, 2013; 19:627–631.

76.

Maamari

, Ausayakhun

, Margolis

, et al. Novel telemedicine device for diagnosis of corneal abrasions and ulcers in resource-poor settings. JAMA Ophthalmol, 2014; 132:894–895.

77.

Jirawison

, Yen

, Leenasirimakul

, et al. Telemedicine screening for cytomegalovirus retinitis at the point of care for human immunodeficiency virus infection. JAMA Ophthalmol, 2015; 133:198–205.

78.

Lee

, Lee

, Nasir

, et al. Diabetes telemonitoring reduces the risk of hypoglycaemia during Ramadan: A pilot randomized controlled study. Diabet Med, 2015; 32:1658–1661.

79.

Matsumoto

, Vongsa

, Nakajima

, et al. Assessment of quality and interpretation of panoramic radiographs obtained in the Lao People's Democratic Republic as part of a teleradiology collaboration with Japan. J Oral Sci, 2015; 57:235–239.

80.

Lim

, Lee

, Chua

, et al. Remote monitoring of patients with cardiac implantable electronic devices: A Southeast Asian, single-centre pilot study. Singapore Med J, 2016; 57:372–377.

81.

Nguyen

, Tan

GSW

, Tapp

, et al. Cost-effectiveness of a national telemedicine diabetic retinopathy screening program in Singapore. Ophthalmology, 2016; 123:2571–2580.

82.

Chinprasatsak

, Satthapong

, Kotruchin

, Maporn

. Telegraphic medicine systems improve medical diagnosis in pre-hospital settings: A pilot study in a tertiary care hospital. J Med Assoc Thai, 2017; 100:686–691.

83.

, Rashid

, O'Brien

. Determining the effectiveness of a video-based contact intervention in improving attitudes of Penang primary care nurses towards people with mental illness. PLoS One, 2017; 12:e0187861.

84.

Xie

, Nguyen

, Hamzah

, et al. Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: An economic analysis modelling study. Lancet Digit Health, 2020; 2:e240–e249.