Telemedicine interventions in type 2 diabetes management: a protocol for systematic review and network meta-analysis

Abstract

Introduction

The consequences of type 2 diabetes mellitus (T2DM) heavily strain individuals and healthcare systems worldwide. Interventions via telemedicine have become a potential tactic to tackle the difficulties in effectively managing T2DM. However, more research is needed to determine how telemedicine interventions affect T2DM management. This study sets out to systematically analyse and report the effects of telemedicine treatments on T2DM management to gain essential insights into the potential of telemedicine as a cutting-edge strategy to improve the outcomes and care delivery for people with T2DM.

Methods and analysis

To uncover relevant research, we will perform a comprehensive literature search across six databases (PubMed, IEEE, EMBASE, Web of Science, Google Scholar and Cochrane Library). Each piece of data will be extracted separately, and any discrepancies will be worked out through discussion or by a third reviewer. The studies included are randomised controlled trial. We chose by predefined inclusion standards. After the telemedicine intervention, glycated haemoglobin will be the primary outcome. The Cochrane risk-of-bias approach will be used to evaluate the quality of the included studies. RevMan V.5.3.5 software and RStiduo V.4.3.1 software can be used to analyse the data, including publication bias.

Ethics and dissemination

Since this research will employ publicly accessible documents, ethical approval is unnecessary. The review is registered prospectively on the PROSPERO database. The study’s findings will be published in a peer-reviewed journal.

PROSPERO registration number

CRD42023421719.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Network meta-analysis is suitable for multi-intervention comparisons.

• Subgroup analysis will be used to adjust the presence of heterogeneity. However, there is a concern about the need for more sufficient subjects and studies for a practical analysis.

• Grey literature will not be included to ensure the rigour of the results as per our protocol.

Introduction and description of the condition

Telemedicine has evolved significantly since the early 2000s, fuelled by technological advancements and the global COVID-19 pandemic. The first phase of telemedicine primarily involved using communication to deliver medical services remotely, as defined by the WHO, and has grown and evolved significantly in the twenty-first century.^{1 2} Early telemedicine apps primarily videoconferencing and remote monitoring, with studies demonstrating their feasibility and effectiveness in various medical fields, such as cardiology, dermatology and mental health.^3–5 The emergence of mobile health (mHealth) apps, remote diagnostics and virtual reality, among other developments, marked a change in the second phase of telemedicine’s technological environment.^6–10 Integrating electronic health records during this phase made remote medical service delivery easier, individual outcomes improved and healthcare expenditure cost-effectively reduced.^{10 11} Through remote tracking and self-management, studies have shown that mHealth devices have successfully managed chronic conditions like diabetes and hypertension.^10–12 The COVID-19 pandemic has significantly accelerated the adoption and use of telemedicine globally in the third phase, highlighting its potential to improve access to treatment, particularly for underserved areas.¹³ The COVID-19 problem, which surfaced in 2019, hastened telemedicine’s integration into traditional healthcare systems, however, and underlined its critical role in providing continuity of care in the face of physical segregation policies and overburdened healthcare infrastructures. In the post-COVID era, The WHO has recognised telemedicine as an essential tool in response to the pandemic.⁹ Telemedicine is anticipated to play a significant part in healthcare provision and provide even more tremendous potential for individual outcomes improvement and cost reduction.

Due to technological advancements and the rising demand for effective healthcare services, telemedicine has become crucial in managing and treating non-communicable diseases (NCDs).¹⁴ The term ‘NCDs’ refers to chronic, non-infectious diseases that various things can bring on, including genetics, way of life and environmental variables. Heart disease, cancer, diabetes and persistent respiratory conditions are a few examples of NCDs. Telemedicine’s application in the treatment of NCDs has produced encouraging results for improving individual outcomes and lowering medical expenses. Diabetes is global, and more than 95% of people with diabetes have type 2 diabetes mellitus (T2DM), which makes up 90% of all diabetes cases.¹⁵ Effective diabetes treatment strategies frequently combine self-management with ongoing support from healthcare professionals. The complexity of diabetes leads to difficulty in self-management. Individuals must make significant lifestyle changes, such as adjusting their diet and exercise regimens, routinely monitoring their blood glucose levels and correctly using prescribed medications to achieve effective self-management.¹⁶ Several strategies to improve care during glycaemic management are generally individual based, which requires more clinic visits, frequent monitoring of outcomes, providing diabetic and dietetic education, promoting self-management, and finally, escalation of injectable medications like insulin titration. Diabetes can have severe micro and macrovascular complications.¹⁷ Additionally, individuals with diabetes are more susceptible to poor outcomes for infectious diseases, including COVID-19, as found by the WHO.¹⁵ Given the severity of the consequences associated with diabetes, evaluating and analysing the effectiveness of interventions such as telemedicine for managing T2DM in primary healthcare is crucial.^{18 19}

The main elements during telemedicine in diabetic individuals primarily focus on close blood sugar monitoring through frequent virtual follow-up, titration of insulin doses, managing hypoglycaemia episodes, reviewing recent laboratory results, ordering medication refills and suggesting treatment plan modifications when needed. This strategy can enhance individual outcomes, boost happiness and lower healthcare expenses for controlling T2DM.²⁰ Telemedicine is proposed as an innovative solution to enhance diabetic patient care, offering services such as disease prevention information, medical condition monitoring and regular diabetes management.²¹ It enables accurate data communication between patients and healthcare providers, empowering patients towards healthier lifestyles and better glycaemic control. These services can be synchronous, asynchronous or continuous.^22–24 Despite its potential benefits, implementing telemedicine in conventional care is complex and faces resistance from some patients.²⁵ Limited studies in low to middle-income countries like Malaysia highlight a gap in understanding patient perspectives, as existing research primarily focuses on physician views.^26–30 Successful implementation research should address the behaviour of stakeholders at various levels, especially in resource-constrained settings.

This research aims to examine the effects of telemedicine interventions on clinical outcomes in individuals with T2DM by measuring glycaemic control by including glycated haemoglobin (HbA1c) levels as the primary outcome. The secondary outcome is to measure the impact of telemedicine conducted by a different delivery mechanism on blood pressure, fasting blood sugar, lipid profile and body weight. Also, there are drawbacks and current restrictions to the remote provision of primary healthcare for treating this condition. Especially for those with limited access to care, telemedicine is being accepted as a promising approach to increase the accessibility, quality and efficiency of healthcare services. The results of this study could offer crucial insights into the potential advantages and restrictions of telemedicine for the management of T2DM and add to the expanding body of literature on remote healthcare delivery for the management of chronic diseases. Building on existing research conducted in the same domain, the findings of this study will contribute to the growing body of knowledge on the effectiveness and limitations of telemedicine interventions in managing T2DM.^31–33 The release of a study in May 2023 proves that this study considers the most recent research.³⁴ Nevertheless, it is essential to note that before this publication, the research protocol was designed, and funding was obtained in 2022 to start the study. This study offers two advantages: first, a subgroup analysis is included to examine how COVID-19 has affected telemedicine interventions, emphasising T2DM management. This analysis provides insightful information about the pandemic’s particular effects. Second, by including the variable of the infrastructure development index in the study, it is possible to investigate how changes in healthcare infrastructure affect the efficacy of telemedicine interventions.

Systematic review questions

• What are telemedicine methods implemented in managing T2DM?

• Does telemedicine impact HbA1c levels in T2DM compared with standard care?

• What telemedicine method has the highest effect on HbA1c level?

• What are the advantages, disadvantages and potential limitations of telemedicine methods?

In adherence to the Cochrane framework, our methodology employs the Population, Intervention, Comparison, Outcome format to establish a systematic approach to inquiry. Our study focuses on individuals afflicted with T2DM. Specifically, the intervention under investigation centres on the utilisation of telemedicine, which manifests through diverse mediums, including but not limited to telephonic outreach, short message service correspondences, mobile applications encompassing features for both voice and video communication, website-based interfaces and the integration of internet of things smart devices. This assortment of interventions includes integrating wearable technologies such as pedometers and heart rate monitors. In the comparative analysis, we juxtapose our findings against the backdrop of standard care procedures prevalent within the healthcare domain.

As for the primary outcome of our investigation, we scrutinise the alteration in levels of HbA1c test used routinely to evaluate the glycaemic control of diabetes individuals after individuals receiving telemedicine care, and routinely, the HbA1c is measured every 3 months. HbA1c goals must be adjusted based on the agreed goal, considering the individual’s age and the presence of comorbidities associated with diabetes. However, they should always be kept under non-diabetic values seven and around.³⁵ Numerous routines, such as exercise, diet programmes, insulin administration and dose titration and self-blood glucose monitoring, must be followed by individuals. Additionally, our secondary analysis extends to encompass an evaluation of subgroups differentiated by metrics such as blood pressure, denoted in millimetres of Mercury (mm Hg), body mass index (BMI), measured in kilograms per square metre (kg/m²), and low-density lipoprotein (LDL) cholesterol levels, indicated in millimoles per litre (mmol/L).

Methods and analysis

Eligibility criteria

We adhered to the 2015 Preferred Reporting for Systematic Review and Meta-Analysis Protocols Project reporting requirements to create this protocol. Our meta-analysis will include published randomised controlled trials (RCTs) of telemedicine on T2DM individuals. Figure 1 displays a flowchart of the study selection process. Patient and public involvement are not applicable in our study as it is a systematic review and meta-analysis of published data.

Figure 1.

Overview of study selection process adhering to PRISMA guidelines. PRISMA, Preferred Reporting for Systematic Review and Meta-Analysis.

Inclusion criteria

The inclusion criteria include RCTs involving T2DM individuals that have HbA1c as test value received a telemedicine intervention via telemonitoring, internet telephone and applications and are among the inclusion criteria and restricted to publications in the English language. The search process will be restricted to RCTs published after 2009. The Data sources describes the individuals and interventions’ criteria in more detail.

Exclusion criteria

The exclusion criteria included studies that used diabetes type 1, did not have HbA1c as a primary outcome and used subgroups from original studies, grey literature, and case studies and reports.

Data sources

We will use PubMed, EMBASE, Web of Science, Cochran, Google Scholar and IEEE library databases for the systematic search of individual studies, with inclusion criteria such as interactions between individuals and clinicians (doctor, nurse or another allied health professional), articles in English, original research studies, the studies report at least one clinical outcome per the definition used in the protocol. The study should report sample size or effect size, such (as the difference between means, proportions, OR, relative risk and hazard risk). The search process started on the 2 January 2023 and is scheduled to end by 30 July 2023. The results did not include reports and grey literature but were reviewed by experts for further insight.

Search strategy

We developed a search query for PubMed, EMBASE, Web of Science, Cochran, Google Scholar and IEEE databases. These databases were picked since they gather much data from medical field studies. Furthermore, when creating query strings, we used the PubMed guidelines.³⁶ All relevant terms are included in the query string and then connected using AND/OR operators. All incorporated terms, synonyms and variations are related using the OR operator. In contrast, the primary terms are linked using the AND operator to connect two ideas about using a * at the root of a word to find numerous endings. Consultants were involved in the query revision to ensure it was thorough and accurate. In April 2023, the query string was used and applied across all databases we mentioned. The entire proposed search term is: (“Telemed*” OR “e-medicine” OR “e-health” OR “tele-health” OR “m-health” OR “mhealth” OR “m* health” OR “mobile-health” OR “Tele-monitoring” OR “*ecare” OR “Diabetes Self-management education and support” OR “Remote Consultation” OR “internet-based” OR “web-based” OR “online” OR “digital” OR “web” OR “phone” OR “video” OR “electronic” OR “remote” OR “virtual” OR “mobile”) AND (“T2DM” OR (diabetes mellitus type 2) OR “diabetes mellitus” OR “T2DM efficacy” OR “glycaemic control” OR “HbA1c” OR “pregnancy diabetes mellitus” OR “gestational diabetes” OR “type 2” OR “gestational” OR “impaired glucose” OR “insulin resistance”) AND (“Clinical Effectiveness” OR “Treatment Effectiveness” OR “Patient-Relevant Outcome?” OR “Patient Relevant Outcome” OR “Clinical Efficacy” OR “Treatment Efficacy” OR “Rehabilitation Outcome”)

Therefore, “And” will concatenate the key terms during the search. We will manually search and cross-reference the reference lists of the gathered studies to locate potentially pertinent articles. The methodology incorporates a meticulously structured process to ensure a comprehensive and rigorous approach to search results. Two primary reviewers, WM and OA, will independently develop initial drafts of the search terms, working in isolation to foster diverse perspectives and mitigate the potential influence of group bias. This dual approach promotes the generation of a comprehensive array of potential search terms and strategies, thereby enhancing the scope of the search process. Following the individual formulation of search terms, a collaborative meeting will convene, where both reviewers will present their respective search strategies and terminologies. This session will facilitate a detailed discussion to identify areas of concurrence and discordance between the two approaches. In cases where a consensus on the search terms is achieved, ensuring alignment and clarity, the words will be adopted without further contention. However, if divergent opinions emerge between the two reviewers, a resolution mechanism will be activated. A decisive vote will be cast by a designated third reviewer, AT, who will serve as an impartial arbiter to resolve conflicts of opinion between the primary reviewers. This approach fosters an unbiased and equitable decision-making process, thereby safeguarding the integrity and impartiality of the search and review procedures. The involvement of the third reviewer, AT, as a final arbiter, underscores the commitment to maintaining transparency and adherence to robust scientific methodologies, ultimately ensuring the reliability and validity of the research findings.

Study selection

After removing the other publications, the complete texts of the RCT studies will be retrieved for further analysis. After eliminating duplicates, we will review the individual literature, reject publications that do not meet the requirements, and note the reasons for exclusion. After that, possibly eligible literature will be thoroughly examined to eliminate any remaining ineligible content and to record why it was excluded. If we suppose the literature cannot be adequately incorporated, in that case, it will be rescreened, which may entail contacting the authors via phone, mail or other methods to gather crucial data.

Data extraction

We will develop a standard data extraction form that contains the following details: title, author’s name, publication year, number of participants (sample size), general application, specific application, intervention, other intervention if applicable, control group (standard care) in all studies, country, gender, age, weight, BMI, blood pressure, demographics, HbA1c (study outcome), the duration of follow-up, self-efficacy score, functionality, medication, mental health depression, chronic diabetes, current alcohol user, smoking history, cardiovascular disease (CVD) risk, reliant on insulin or not, complete cholestatic, fasting plasma glucose, postprandial glucose and glucose monitoring, comorbidity LDL, HDL, triglycerides, lipids, dyslipidaemia, quality of life and complications. We will each independently extract and input data from the papers onto a data sheet. A third researcher (MA) will examine the results of all the studies. If there is a disagreement over the data provided, we will discuss it and find a solution; if required, the third investigator will join the conversation. In addition, if necessary, inquiries will be directed to the authors if there are any missing, unclear or insufficient data in the manuscript; otherwise, the study will not be considered.

Quality assessment

The Cochrane risk of bias assessment tool will ensure that the studies are thoroughly assessed for quality. This tool examines biases such as creating random sequences, hiding allocations, subject and researcher blindness, insufficient outcome data and selective reporting. A grade of ‘low bias risk’, ‘bias uncertainty’ or ‘high bias risk’ will be given to each indicator.

Outcome measures

RevMan V.5.4 will be used to construct this tool and R language³⁷ on RStudio V.4.3.1 software.³⁸ This method will allow for a structured and consistent evaluation of study-related factors like selection, performance, detection, attrition, reporting and other bias sources. The review will offer a thorough and impartial assessment of the quality of the included studies by employing this method.

Data synthesis

The synthesis of data is designed to encompass both qualitative and quantitative methodologies. The qualitative component entails a systematic review and comprehensive assessment of the available RCTs. This process will involve meticulous scrutiny of individual studies, with the evidence synthesis predicated on predefined criteria outlined within the established protocol. Conversely, the quantitative aspect of the synthesis will rely on the implementation of network meta-analysis models. Given the diverse nature of telemedicine, which encompasses a spectrum of methods, the network meta-models will facilitate both direct and indirect comparisons, particularly when contrasted with the conventional framework of diabetes care. In constructing the meta-model and choosing the appropriate effect, the meta-model assumes critical significance. Initially, a fixed-effect model will be applied, followed by exploring subgroups within the fixed-effect model, using age as an illustrative example. In cases where the assessment of heterogeneity indicates a necessity for alternate modelling, the random-effect model will be considered. The decision to select the appropriate model will be contingent on rigorous inspections of heterogeneity.

Subgroup analysis

First, CVD is widely recognised as the primary cause of mortality and morbidity in individuals with T2DM. Second, we will investigate the impact of physical activity and lifestyle Interventions. Third, we will examine countries’ infrastructure, focusing on the necessary elements to provide effective diabetes care in a given nation. These variables will be measured using proxy indicators. The country where the RCT is conducted, the country’s Human Development Index (HDI), the country’s development index according to the World Bank and other indicators will be assessed. Furthering the initial two points, we will extract standardised binary outcomes, specifically discerning the prevalence or absence of CVD within the individuals recruited for each RCT. This systematic classification will serve as a cornerstone in fostering a comprehensive understanding of the disease’s prevalence within the studied cohort. Similarly, this approach will be applied to other potential covariates, such as physical activity. Other counts and continuous covariates, such as the average age, number of comorbidities, average LDL, and others, will be included as necessary in a meta-regression model.

Additionally, the HDI evaluation will be meticulously structured to accommodate varying scores, thereby enabling a comprehensive categorisation of nations based on their relative developmental statuses. This stratified approach will categorise countries into distinct groups, denoting levels of HDI such as very high, high, medium and low, consequently facilitating a nuanced analysis of the developmental landscape of these countries and its potential implications for the dynamics of diabetes care and implementation of telemedicine. Concurrently, incorporating subgroup analyses will be pivotal in delineating and comprehensively evaluating discrete segments within the research cohort.

Patient and public involvement

Patient and public involvement are not applicable in our study as it is a systematic review and meta-analysis of published data.

Discussion

This study aims to conduct a systematic review and meta-analysis to assess how telemedicine treatments affect individuals with T2DM. Comorbidities and harmful effects of T2DM have been observed to decrease when HbA1c levels are well controlled.³⁹ However, many individuals require assistance obtaining and maintaining good HbA1c management because of factors like cost-effectiveness, time constraints and distance, especially in isolated communities or regions with few healthcare resources.⁴⁰ As a result, a sizeable fraction of people with critical T2DM do not receive treatment or have HbA1c levels that are out of control, which adds to the overall burden of the disease.⁴¹

Interventions in telemedicine that enable remote healthcare delivery and support present a viable remedy. These interventions use various technologies to provide individuals with virtual consultations, telephone or email follow-ups, health education, nutrition and exercise advice, medication management and reminders of their treatment plans. Individuals can improve their self-management, gain remote access to medical professionals and manage their diabetes more effectively overall by utilising telemedicine.⁴²

Nevertheless, despite the potential advantages of telemedicine in managing T2DM, further research is still required to assess its impact on individual outcomes and clinical efficacy. Little research on the effectiveness of telemedicine interventions in managing T2DM has frequently concentrated on particular components or therapies. The best telemedicine interventions for various subgroups of individuals with critical T2DM are still to be determined as is the best way to implement them.

This systematic review and meta-analysis will thoroughly assess the effects of various telemedicine interventions on individuals with critical T2DM to close these gaps. RCTs will be used in the study to examine remote T2DM therapies that involve individual follow-up and other features.

Decreasing or controlling HbA1c levels will be the study’s primary performance indicator. The study will also undertake subgroup analyses to investigate further the potential effects of variables including country, age, HbA1c classification and the type of telemedicine intervention employed on individual outcomes.

It is vital to recognise any restrictions that might impact the outcomes of this meta-analysis. In addition to differences in sample size, participant characteristics and the specific elements and implementation methods of telemedicine interventions across the included research, these constraints have publication bias, which may cause the exclusion of relevant studies. This study’s heterogeneity will be appropriately considered during data analysis.

This study seeks to provide thorough and impartial information regarding the impact of telemedicine therapy on individuals with T2DM by using stringent search strategies and statistical analysis. The results will help us better understand how successful telemedicine interventions are, the proportion of different types of telemedicine and a contribution to the body of evidence to drink and support health policies and T2DM management locally and globally.

Ethics and dissemination

This review, rooted in examining publicly available documents without requiring direct human participation, obviates the need for ethical clearance. The pre-registration of our research protocol on the PROSPERO database underscores the meticulous adherence to methodological transparency. The dissemination strategy for this study involves publication in a peer-reviewed journal. Crucially, this review is an integral component of a broader investigation into telemedicine and diabetes control, a project that receives financial support from Princess Sumaya University for Technology (fund number: 134/38/31/4). This funding bolsters our commitment to advancing research in this vital domain. Our dissemination approach, encompassing PROSPERO registration and subsequent peer-reviewed publication, aligns with the highest standards of scholarly communication, ensuring the accessibility and transparency of our research findings.

Ethics statements

Patient consent for publication

Not applicable.